Autobooks is a Detroit-based fintech company that partners with financial institutions (FIs) to provide an integrated invoicing, payments, and accounting solution for small businesses delivered through online banking

Autobooks is simple banking that works for small businesses. Autobooks works with financial institutions to deliver what small businesses need: cash flow management, payment services, and financial reporting integrated directly into current banking channels. Autobooks provides a simple, yet automated approach to small business banking. Our solution makes accounting and payments easy to understand, which helps small business in their day-to-day, giving them more time to do what they do best. Autobooks creates opportunities for established business owners, as well as emerging micro, lifestyle and gig economy entrepreneurs.

Autobooks

autobooks

We're Just Riding the Ups and Downs of Investing in Bitcoin

HackerNoon Reporter: Please tell us briefly about your  background. I am a ‘late start’ female founder from Ghana after a 25+ years successful career in IT for Development, Service Delivery and Content Creation. I am fuelled by a passion for books & the immense potential of the internet and mobile phones for spreading African knowledge. What's your startup called? And in a sentence or two, what does it do? AkooBooks Audio is a platform and experience provider focused on transforming Black/African books into engaging audiobooks that reach new audiences on mobile phones. Our library of African audio is available anytime, accessible anywhere on a wide variety of mobile devices at flexible subscription plans and low retail cost in various Ghanaian and African languages and not only in English. What is the origin story? The seeds for my business were first inspired by my mother, Nana Dadson, a children's book author and avid reader. My mother lost her vision over a decade ago and unfortunately with that her ability to read African literature independently, as only a fraction is available in digital form. Soon, AkooBooks Audio was born. Akoobooks Audio What do you love about your team, and why are you the ones to solve this problem? We are creating a literate social environment that goes beyond equipping schools with textbooks and libraries and enabling  learners to use their knowledge appropriately and meaningfully in everyday life. Shortage of reading material for effective literacy, limited know-how and slow adoption of digital book distribution prevent efficient and effective African knowledge transfer and stunt our growth in the digital economy. With Africa recording the highest penetration of the use of mobile phones, the potential of audiobooks  and digital reading comes in. Our team has over 30 years combined experience in technology, publishing, entertainment and media. If you weren’t building your startup, what would you be doing? Mothering my teenager and planning how to open up a restaurant/cafe. At the moment, how do you measure success? What are your core metrics? We look at the number of titles in our catalogue and  currently run a retail per streaming download model. As a digital publishing company, subscriptions are fast becoming a a way of competing with other online entertainment models that are free. We want to move to an all you can eat streaming model. Mobile platform offers new sources to inform decision making, we share  data with publishers to help them make the right production choices for future catalogues. Our key performance indicators (KPIs) for our audiobooks include: a) Relevance- how many users start listening to an audiobook; b)Drop off rate- how many users stop listening to an audiobook c) Finish rate-how many users finish the book to the end. Also, data on preferences, Narrators, Genres and Barriers. What’s most exciting about your traction to date? We’ve won 3 business awards that  have enabled us to roll out our web based African audiobook platform which is  also available on the Google and Apple playstores. We have licensed over 300 African titles from African publishers all over the African continent. We hope to partner the Ministry of Education to bring digital books into Ghanaian classrooms. What technologies are you currently most excited about, and most worried about? And why? Audio and AI/machine technology. Our programme promotes audio literacy and pilots it in local languages, bringing a wealth of ideas and experiences to people who are illiterate in English. We see the opportunity to skip the translation phase of books and go directly from English audio into producing ‘straight into audio’ local language versions. The new interpreted local language adaptations can then be transcribed in local languages and form the basis for new local language literature for young people. With our audio data, Akoobooks Audio is at the forefront of utilizing African voice interaction and ML technology to connect African books and authors to new readers worldwide. What drew you to get published on HackerNoon? What do you like most about our platform? I have only been following the platform for a few months and pleasantly surprised to find a good coverage of African tech content that is engaging and insightful. What advice would you give to the 21-year-old version of yourself? To be positive and believe in myself. When I encounter failure, take a step back and realize that failure is part of life and to ask myself if there are any lessons to be learned or if there are other ways around it. Being a woman in a male dominated career space like tech or publishing can be challenging, but our femininity is our greatest strength. What is something surprising you've learned this year that your contemporaries would benefit from knowing? Overseas capital is flooding the African fintech and African digital streaming subscription services market for entertainment (movies -Netflix and music -Spotify). I look forward to seeing similar developments in the online reading space! Vote for Accra Startups of the Year!

My Mother Lost Her Vision, So I Made African Books into Engaging Audiobooks

Isn’t it interesting? So basically after reading this article you don’t need to read long PDFs which contains more than 10 pages and only with help of 12 lines Python code. I’ll explain every line of code so even who doesn’t have knowledge of Python, can understand the code. So let’s start!! Here’re the steps to follow for making an AudioBook: First of all you must have Python and it’s IDE installed on your PC If you don’t have these installed you can do it just by clicking on and . There are many IDEs but I’m giving you link of PyCharm because it is easy to use and open-source. If you don’t know IDE than no worry about it’s just a code editor for Python. Python IDE After downloading both Python and PyCharm you have to set it’s environment and for that there are many videos on YouTube so take help from them. Let’s open IDE and create a new file and give it a name whatever you want. Click on Terminal option then just installed Python text to speech library just like this : (pyttsx3) pip install pyttsx3 #make sure you are connected with Internet Also install PyPDF2 library in same way: pip install PyPDF2 : Here is the full code pyttsx3 PyPDF2
pdf_book = ( , )
pdf_book_reader = PyPDF2.PdfFileReader(pdf_book)
pages_no = pdf_book_reader.numPages
print(pages_no)
speaker = pyttsx3.init() number range( ,pages_no):
  page_start = 
  pdf_book_reader.getPage( ) = page_start.extractText()
  speaker.say( )
  speaker.runAndWait() import import open 'book.pdf' 'rb' for in 8 8 text text Explanation of code: In line 1 & 2 import required libraries. In line 3 , create a variable called which is used to open the pdf file which we want to make speak and ‘rb’ is to read in binary. pdf_book In line 4, read the file which we want to make speak using function of library. pdf_book_reader = PyPDF2.PdfFileReader(pdf_book) PdfFileReader PyPDF2 In line 5, counting total number of pages in the file using numPages function. pages_no = pdf_book_reader.numPages In line 6, displays total number of pages in the file. print(pages_no) In line 7, initiate the pyttsx3 library to speak the file. speaker = pyttsx3.init() In line 8, a for loop is initiated to iterate through whole pdf file from starting page to ending page using range function. for number in range(8,pages_no): In line 9, starting page from where we want to make Python to read. page_start = pdf_book_reader.getPage(8) In line 10, text = extracting the text from starting page using function. page_start.extractText() extractText() In line 11, speaking the text from PDF file pages using say function. speaker.say(text) In line 12, making sure that it speaks all pages for PDF file speaker.runAndWait() So now you can lay on your bed and just listen whatever PDF you want to speak anytime, anywhere from Python just by running this program. Note: I’ll suggest you to put the PDF file in the same folder where you are writing code and it’ll reduce your trouble. Thank you so much for reading! follow the writer of the article for more stuff on Python and Data Science.

How To Convert PDFs Into AudioBooks

Young children are always told to read more books to increase their performance at school, as well as help with their cognitive development. Reading books is a hobby that inculcates creativity, vocabulary, patience and concentration, however, in today’s fast-paced world, it might be a little hard to completely devote some time of the day simply to reading. Therefore, today, it has become almost a novelty to own a large collection of books and more importantly, to have the kind of time to actually read them all. This is understandable, of course, because reading is an intensive exercise that requires undivided attention from our minds and at the same time, needs our eyes and ears to be focused on that one aspect only. It is only natural that most people today may not find that kind of time and space in their everyday lives. If find yourself looking for a break from reality from time-to-time to drown in a story, audiobooks may just be what you are looking for. What are audiobooks? In very simple terms, audiobooks are simply voice recordings of stories from physically written books. These are available on numerous apps and websites and can be heard from any digital device like a laptop, smartphone, tablet or iPod. The debate surrounding the nature of audiobooks vs physical books focuses on whether this new innovation aims to take away from the habit of reading. A lot of people who love to read consider audiobooks as a “cheat” escape, meaning that these are not nearly as good as actual books but can perform the same functions. In light of this, it is important to point out some very important ways audiobooks have been more accommodating when it comes to our busy lives. How have audiobooks improved into our lives? have become very popular because they sync so perfectly with our busy lives. However, other than the obvious argument about how convenient it is for people to actually integrate it, there are many other benefits for relying on these, even for young kids. Some of these include: Audiobooks They don't deteriorate your vision One of the main problems with reading from books and tablets is that it gives a lot of respite to the eyes. For a lot of people with vision problems, reading can be very stressful to their eyes, despite reading from devices that do not give out any glare. Audiobooks do not require any visual reading and therefore, they are an ideal choice for people who especially suffer from vision problems. They don't take away from sleeping hours Glare from tablets and other devices can not only affect the eyes but also affect the circadian rhythm, which in turn affects sleep patterns. Therefore, audiobooks will also help you get a restful night of sleep. A lot of people use audiobooks as ASMR as well, since this can lull them into sleeping more easily. Audiobooks can help with time management Since audiobooks are designed to gel perfectly with our daily lives, they do not require people to devote time separately to simply listening to them. These can be listened to at any point of time, while driving, or having a bit of a break at the office, in between classes, while taking a walk or just while lazing around. This helps people who have very little time at hand and therefore cannot indulge in reading because they feel they will squander away whatever little free time they have. Moreover, if audiobooks are enjoyed while doing mundane everyday jobs, the experience can become more enjoyable. For example, you may dislike doing the dishes, but if you make it a point to listen to an audiobook while you do your dishes, you will find that you no longer feel like the task is a compulsion, and you will get it accomplished much faster. Audiobooks can help with pronunciation One of the prime requirements for being an audiobook narrator is having a soothing voice with flawless pronunciation. While this is a relatively less-explored advantage of listening to audiobooks, one has to agree that they help in pronunciation and developing a better vocabulary. This is very useful for young children, who can learn new words and correct pronunciations by listening to audiobooks. Migrants and people learning a second or a third language will also reap the benefits from this since continuously listening to audiobooks will directly help them with learning the language better. Audiobooks are just as helpful as reading physical books Numerous researchers say that the same parts of the brain are stimulated while reading a book or listening to an audiobook. The same cognitive and emotional responses have been noticed irrespective of the kind of medium they are read/listened to. This means recollection of information is the same regardless of the medium. It can be more emotionally intimate A lot of us like it when someone narrates a story to us. In this case, we can be the third person who witnesses the interaction between all the characters. This is why a lot of people connect better to audiobooks than they do with reading. It is a more affordable solution One of the reasons a lot of people steer away from reading is because books are not necessarily cheap, and they also require a lot of space and maintenance. For example, for those who have a hobby of reading, you may invest in a glare-free device or in lots of books, which is not always the most affordable solution. Therefore, audiobooks are an affordable alternative, where you can simply use the device you are already on, without having to buy more books and keep looking for space to store them. Other than these, a lot of people have affirmed that listening to audiobooks has been very helpful for them in dealing with mental health issues like anxiety. Reading a book might feel like a huge task that takes up time, but when it comes to listening to an audiobook, a person can feel that they are multitasking, as they are not necessarily squandering away their time and also engaging in a hobby simultaneously. Therefore, one can see that audiobooks suit our busy lifestyles so that we can enjoy a few minutes of quiet retreat without necessarily feeling unproductive. This in no way implies that reading books should no longer be followed, of course. However, for people who might feel like they are getting distanced from their hobbies, audiobooks can be a great way to rekindle that relationship they once had with stories. About The Author While a lot of websites offer audiobooks in different languages, at , we specifically focus on bringing back the lost taste of Indian classics from the likes of Premchand, Amrita Pritam and so on. We aim to bring Indian literature to all sections of society and our audiobooks range from stories, poems and motivational speeches. Gaatha Also published at https://gaatha850.medium.com/audiobooks-how-have-helped-our-busy-lives-by-gaatha-9aa3de919e2a

Audiobooks Provide The Same Cognitive Recollection as Physical Books Do 

Inspired by the , I’ve decided to share a collection of my favorite audiobooks and podcasts that I think HN readers will love. This is the best stuff I have discovered over the past 3–4 years, and I’m sure you guys will enjoy each of these books, so if you haven’t read some of them yet — I highly recommend you to check them out. recent HN thread Excellent audiobooks _From the founders of the trailblazing software company 37signals, here is a…_www.audible.com Rework _From Ed Catmull, co-founder (with Steve Jobs and John Lasseter) of Pixar…_www.audible.com Creativity, Inc. _Check out this great listen on Audible.com. 2015 Audie Award WINNER, Science Fiction 2015 Audie Award Finalist, Solo…_www.audible.com The Martian _Check out this great listen on Audible.com. Ben Horowitz, cofounder of Andreessen Horowitz and one of Silicon Valley…_www.audible.com The Hard Thing About Hard Things _Check out this great listen on Audible.com. Every moment in business happens only once. The next Bill Gates will not…_www.audible.com Zero to One _Check out this great listen on Audible.com. Jeff Hawkins, the man who created the PalmPilot, Treo smart phone, and…_www.audible.com On Intelligence _Check out this great listen on Audible.com. First, businesses discovered quality as a key competitive edge; next came…_www.audible.com The Design of Everyday Things _Check out this great listen on Audible.com. With his characteristic eyebrow-raising behavior, Richard P. Feynman once…_www.audible.com Surely You're Joking, Mr. Feynman! _Check out this great listen on Audible.com. Yuval Noah Harari, author of the critically acclaimed New York Times best…_www.audible.com Homo Deus _Check out this great listen on Audible.com. Atlas Shrugged is the "second most influential book for Americans today…_www.audible.com Atlas Shrugged _Check out this great listen on Audible.com. Map the innovation space - and blaze a path to profits and growth…_www.audible.com Mapping Innovation _Check out this great listen on Audible.com. Kevin Mitnick, the world's most famous hacker, teaches you easy cloaking…_www.audible.com The Art of Invisibility _Check out this great listen on Audible.com. The new era of gamification and human-focused design optimizes for…_www.audible.com Actionable Gamification _Check out this great listen on Audible.com. A fascinating exploration of how computer algorithms can be applied to our…_www.audible.com Algorithms to Live By _Check out this great listen on Audible.com. One of Silicon Valley's most successful angel investors shares his rules…_www.audible.com Angel _Check out this great listen on Audible.com. The mastermind behind Apple Computer sheds his low profile and steps…_www.audible.com iWoz _Check out this great listen on Audible.com. Most startups don't fail because they can't build a product. Most startups…_www.audible.com Traction Excellent podcasts YCombinator https://soundcloud.com/akharris Indie Hackers https://www.indiehackers.com/podcast Syntax (React and WebDev) https://syntax.fm/ Rationally Writing http://alexanderwales.com/rationally-writing/ Audio version of Harry Potter and the Methods of Rationality http://www.hpmorpodcast.com/ Audio version of Rationality: From AI to Zombies https://from-ai-to-zombies.eu/

Top audiobooks for Startup Founders

Table of Contents A Deep Dive into Voice Cloning with SoftVC VITS and Bert-VITS2 Prepare Dataset Extract from a Song UVR Workflows Preparation for vocal recording Cheapskate’s Audio Equipment Audacity workflows audio-slicer Cleaning dataset Match loudness so-vits-svc Set up environment Initialization Download pre-trained models Dataset Preparation Edit Configs Training Inference Audio Editing so-vits-svc-fork Installation Preparation Training Inference DDSP-SVC Preparation Training Inference Bert-vits2-V2.3 Initialization Preparation Transcription Training and Inference vits-simple-api Tweaks Share models In the , I tried a little bit of and found it interesting. So, I decided to train a usable model with my own voice. previous post TTS Generation WebUI This voice cloning project explores both SVC for Voice Changing and for Text-to-Speech. There is no one tool that does all jobs. VITS I have tested several tools for this project. Many of the good guides, like , , and , are in Chinese. So, I thought it would be useful to post my notes in English. this this this Although has been archived for a few months, probably due to oppression, it is still the tool for the best result. so-vits-svc Other related tools such as , , , and provide either faster/liter optimization, more features, or better interfaces. so-vits-svc-fork so-vits-svc-5.0 DDSP-SVC RVC But with enough time and resources, none of these alternatives can compete with the superior result generated by the original so-vits-svc. For TTS, a new tool called works fantastically and has already matured with its final release last month. It has some very different use cases, for example, audio content creation. Bert-VITS2 Prepare Dataset The audio files of the dataset should be in WAV format, 44100 Hz, 16bit, mono, 1-2 hours ideally. Extract From a Song is the easiest tool for this job. There is a that explains everything in detail. Ultimate Vocal Remover thread UVR Workflows Remove and extract Instrumental Model: VR - UVR(4_HP-Vocal-UVR) Settings: 512 - 10 - GPU Output and unclean vocal Instrumental Remove and extract background vocals. Model: VR - UVR(5_HP-Karaoke-UVR) Settings: 512 - 10 - GPU Output and unclean main vocal background vocal Remove reverb and noise. Model: VR - UVR-DeEcho-DeReverb & UVR-DeNoise Settings: 512 - 10 - GPU - No Other Only Output clean main vocal (Optional) Using RipX (non-free) to perform a manual fine-cleaning. Preparation for Vocal Recording It’s better to record in a treated room with a condenser microphone, otherwise, use a directional or dynamic microphone to reduce noise. Cheapskate’s Audio Equipment The very first time I got into music was during high school, with the blue Sennheiser MX500 and Koss Porta Pro. I still remember the first time I was recording a song that was on a Sony VAIO with Cool Edit Pro. Nowadays, I still resist spending a lot of money on audio hardware as an amateur because it is literally a money-sucking black hole. Nonetheless, I really appreciate the reliability of those cheap production equipment. The core part of my setup is a Behringer UCA202, and it’s perfect for my use cases. I bought it for $10 during a price drop. It is a so-called “Audio Interface” but basically just a sound card with multiple ports. I used RCA to 3.5mm TRS cables for my headphones, a semi-open K240s for regular output, and a closed-back HD669/MDR7506 for monitor output. All three mentioned headphones are under $100 for the normal price. And there are clones from Samson, Tascam, Knox Gear, and more out there for less than $50. For the input device, I’m using a dynamic microphone for the sake of my environmental noises. It is an SM58 copy (Pyle) + a Tascam DR-05 recorder (as an amplifier). Other clones such as SL84c or wm58 would do it too. I use an XLR to 3.5mm TRS cable to connect the microphone to the MIC/External-input of the recorder, and then use an AUX cable to connect between the line-out of the recorder and the input of the UCA202. It’s not recommended to buy an “audio interface” and a dedicated amplifier to replicate my setup. A $10 c-media USB sound card should be good enough. The Syba model that I owned is capable of “pre-amping” dynamic microphones directly and even some lower-end phantom-powered microphones. The setup can go extremely cheap ($40~60), but with UCA202 and DR-05, the sound is much cleaner. And I really like the physical controls, versatility, and portability of my old good digital recorder. Audacity Workflows Although when I was getting paid as a designer, I was pretty happy with Audition. But for personal use on a fun project, Audacity is the way to avoid the chaotic evil of Adobe. Noise Reduction Dereverb Truncate Silence Normalize audio-slicer Use or to slice the audio file into small pieces for later use. audio-slicer audio-slicer (gui) The default setting works great. Cleaning Dataset Remove those very short ones and re-slice those that are still over 10 seconds. In case of a large dataset, remove all that are less than 4 sec. In the case of a small dataset, remove those only under 2 sec. If necessary, perform a manual inspection for every single file. Match Loudness Use Audacity again with ; 0db should do it. Loudness Normalization so-vits-svc Set Up the Environment Virtual environment is essential to run multiple Python tools inside one system. I used to use VMs and Docker, but now, I found that is way quicker and handier than the others. anaconda Create a new environment for so-vits-svc, and activate it. conda create -n so-vits-svc python=3.8
conda activate so-vits-svc Then, install requirements. git clone https://github.com/svc-develop-team/so-vits-svc
cd so-vits-svc

pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118

#for linux
pip install -r requirements.txt

#for windows
pip install -r requirements_win.txt
pip install --upgrade fastapi==0.84.0
pip install --upgrade gradio==3.41.2
pip install --upgrade pydantic==1.10.12
pip install fastapi uvicorn Initialization Download pre-trained models pre-train wget https://huggingface.co/WitchHuntTV/checkpoint_best_legacy_500.pt/resolve/main/checkpoint_best_legacy_500.pt wget https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/rmvpe.pt logs/44k wget https://huggingface.co/datasets/ms903/sovits4.0-768vec-layer12/resolve/main/sovits_768l12_pre_large_320k/clean_D_320000.pth wget https://huggingface.co/datasets/ms903/sovits4.0-768vec-layer12/resolve/main/sovits_768l12_pre_large_320k/clean_G_320000.pth logs/44k/diffusion wget https://huggingface.co/datasets/ms903/Diff-SVC-refactor-pre-trained-model/resolve/main/fix_pitch_add_vctk_600k/model_0.pt (Alternative) wget https://huggingface.co/datasets/ms903/DDSP-SVC-4.0/resolve/main/pre-trained-model/model_0.pt (Alternative) wget https://huggingface.co/datasets/ms903/Diff-SVC-refactor-pre-trained-model/blob/main/hubertsoft_fix_pitch_add_vctk_500k/model_0.pt pre-train/nsf_hifigan wget -P pretrain/ https://github.com/openvpi/vocoders/releases/download/nsf-hifigan-v1/nsf_hifigan_20221211.zip unzip -od pretrain/nsf_hifigan pretrain/nsf_hifigan_20221211.zip Dataset Preparation Put all Prepared audio.wav files into dataset_raw/character cd so-vits-svc
python resample.py --skip_loudnorm
python preprocess_flist_config.py --speech_encoder vec768l12 --vol_aug
python preprocess_hubert_f0.py --use_diff Edit Configs The file is located at configs/config.json : the frequency of printing log : the frequency of saving checkpoints : total steps : numbers of saved checkpoints, 0 for unlimited. : fp32 In my case, : the smaller the faster (rougher), the larger the slower (better). log interval eval interval epochs keep ckpts half_type batch_size Recommended batch_size per VRAM: 4=6G；6=8G；10=12G；14=16G；20=24G Keep default for configs/diffusion.yaml Training python cluster/train_cluster.py --gpu
python train_index.py -c configs/config.json
python train.py -c configs/config.json -m 44k
python train_diff.py -c configs/diffusion.yaml On training steps: Use to train the main model; usually, 20k-30k would be usable, and 50k and up would be good enough. This can take a few days depending on the GPU speed. train.py Feel free to stop it by , and it will be continued training by re-run anytime. ctrl+c python train.py -c configs/config.json -m 44k Use to train the diffusion model; training steps is recommended at 1/3 of the main model. train_diff.py Be aware of over-training. Use to monitor the plots to see if it goes flat. tensorboard --logdir=./logs/44k Change the from 0.0001 to 0.00005 if necessary. learning rate When done, share/transport these files for inference. config/ config.json diffusion.yaml logs/44k feature_and_index.pkl kmeans_10000.pt model_0.pt G_xxxxx.pt Inference It’s time to try out the trained model. I’d prefer Webui for the convenience of tweaking the parameters. But before firing it up, edit the following lines in for LAN access: webUI.py os.system("start http://localhost:7860")
app.launch(server_name="0.0.0.0", server_port=7860) Run ; then access its from a web browser. python webUI.py ipaddress:7860 The webui has no English localization, but would be helpful. Immersive Translate Most parameters would work well with the default value. Refer to and to make changes. this this Upload these 5 files: and its main model.pt config.json and its diffusion model.pt diffusion.yaml Either cluster model for speaking or feature retrieval for singing. kmeans_10000.pt feature_and_index.pkl is for speaking only, not for singing. Recommend when using. F0 predictor RMVPE is useful when singing a feminine song using a model with a masculine voice, or vice versa. Pitch change is the way of controlling the tone. Use to get the clearest speech, and use to get the closest tone to the model. Clustering model/feature retrieval mixing ratio 0.1 0.9 should be set around , it enhances the result at steps. shallow diffusion steps 50 30-100 Audio Editing This procedure is optional. Just for the production of a better song. I won’t go into details about this since the audio editing software, or so-called DAW (digital audio workstation), that I’m using is non-free. I have no intention to advocate proprietary software even though the entire industry is paywalled and closed-source. Audacity supports multitrack, effects, and a lot more. It does load some advanced VST plugins as well. It’s not hard to find tutorials on mastering songs with Audacity. Typically, the mastering process should be mixing/balancing, EQ/compressing, reverb, and imaging. The more advanced the tool is, the easier the process will be. I’ll definitely spend more time on adopting Audacity for my mastering process in the future, and I recommend everyone do so. so-vits-svc-fork This is a so-vits-svc with real-time support, and the models are compatible. Easier to use but does not support the Diffusion model. For dedicated real-time voice changing, a is recommended. fork voice-changer Installation conda create -n so-vits-svc-fork python=3.10 pip
conda activate so-vits-svc-fork

git clone https://github.com/voicepaw/so-vits-svc-fork
cd so-vits-svc-fork

python -m pip install -U pip setuptools wheel
pip install -U torch torchaudio --index-url https://download.pytorch.org/whl/cu118
pip install -U so-vits-svc-fork
pip install click
sudo apt-get install libportaudio2 Preparation Put dataset .wav files into so-vits-svc-fork/dataset_raw svc pre-resample
svc pre-config Edit a in . This fork is a larger size than the original. batch_size configs/44k/config.json Training svc pre-hubert
svc train -t
svc train-cluster Inference Use GUI with . This requires a local desktop environment. svcg Or use CLI with for real time and for generating. svc vc svc infer -m "logs/44k/xxxxx.pth" -c "configs/config.json" raw/xxx.wav DDSP-SVC requires fewer hardware resources and runs faster than so-vits-svc. It supports both real-time and diffusion models (Diff-SVC). DDSP-SVC conda create -n DDSP-SVC python=3.8
conda activate DDSP-SVC

git clone https://github.com/yxlllc/DDSP-SVC
cd DDSP-SVC

pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118
pip install -r requirements.txt Refer to the for the two files: Initialization section pretrain/rmvpe/model.pt
pretrain/contentvec/checkpoint_best_legacy_500.pt Preparation python draw.py
python preprocess.py -c configs/combsub.yaml
python preprocess.py -c configs/diffusion-new.yaml Edit configs/ batch_size: 32  (16 for diffusion)
cache_all_data: false
cache_device: 'cuda'
cache_fp16: false Training conda activate DDSP-SVC
python train.py -c configs/combsub.yaml
python train_diff.py -c configs/diffusion-new.yaml

tensorboard --logdir=exp Inference It’s recommended to use since it includes both DDSP and diffusion model. main_diff.py python main_diff.py -i "input.wav" -diff "model_xxxxxx.pt" -o "output.wav" Real-time GUI for voice cloning: python gui_diff.py Bert-vits2-V2.3 This is a TTS tool that is completely different from everything above. By using it, I have already created several audiobooks with my voice for my parents, and they really enjoy it. Instead of using the , I used the fork by for an easier setup. original v3u Initialization conda create -n bert-vits2 python=3.9
conda activate bert-vits2

git clone https://github.com/v3ucn/Bert-vits2-V2.3.git
cd Bert-vits2-V2.3

pip install torch==2.0.1+cu118 torchvision==0.15.2+cu118 torchaudio==2.0.2+cu118 --index-url https://download.pytorch.org/whl/cu118
pip install -r requirements.txt Download pre-trained models (includes Chinese, Japanese, and English): wget -P slm/wavlm-base-plus/ https://huggingface.co/microsoft/wavlm-base-plus/resolve/main/pytorch_model.bin
wget -P emotional/clap-htsat-fused/ https://huggingface.co/laion/clap-htsat-fused/resolve/main/pytorch_model.bin
wget -P emotional/wav2vec2-large-robust-12-ft-emotion-msp-dim/ https://huggingface.co/audeering/wav2vec2-large-robust-12-ft-emotion-msp-dim/resolve/main/pytorch_model.bin
wget -P bert/chinese-roberta-wwm-ext-large/ https://huggingface.co/hfl/chinese-roberta-wwm-ext-large/resolve/main/pytorch_model.bin
wget -P bert/bert-base-japanese-v3/ https://huggingface.co/cl-tohoku/bert-base-japanese-v3/resolve/main/pytorch_model.bin
wget -P bert/deberta-v3-large/ https://huggingface.co/microsoft/deberta-v3-large/resolve/main/pytorch_model.bin
wget -P bert/deberta-v3-large/ https://huggingface.co/microsoft/deberta-v3-large/resolve/main/pytorch_model.generator.bin
wget -P bert/deberta-v2-large-japanese/ https://huggingface.co/ku-nlp/deberta-v2-large-japanese/resolve/main/pytorch_model.bin Create a character model folder mkdir -p Data/xxx/models/ Download base models: !wget -P Data/xxx/models/ https://huggingface.co/OedoSoldier/Bert-VITS2-2.3/resolve/main/DUR_0.pth
!wget -P Data/xxx/models/ https://huggingface.co/OedoSoldier/Bert-VITS2-2.3/resolve/main/D_0.pth
!wget -P Data/xxx/models/ https://huggingface.co/OedoSoldier/Bert-VITS2-2.3/resolve/main/G_0.pth
!wget -P Data/xxx/models/ https://huggingface.co/OedoSoldier/Bert-VITS2-2.3/resolve/main/WD_0.pth

#More options
https://openi.pcl.ac.cn/Stardust_minus/Bert-VITS2/modelmanage/model_filelist_tmpl?name=Bert-VITS2_2.3%E5%BA%95%E6%A8%A1
https://huggingface.co/Erythrocyte/bert-vits2_base_model/tree/main
https://huggingface.co/OedoSoldier/Bert-VITS2-2.3/tree/main Edit by replacing all to train_ms.py bfloat16 float16 Edit for LAN access: webui.py webbrowser.open(f"start http://localhost:7860")
app.launch(server_name="0.0.0.0", server_port=7860) Edit for and Data/xxx/config.json batch_size spk2id Preparation Similar workflow as in the . previous section Remove noise and silence, normalization, then put the un-sliced WAV file into . Data/xxx/raw Edit for , , and . config.yml dataset_path num_workers keep_ckpts Run to slice the WAV file. python3 audio_slicer.py Clean the dataset ( ) by removing small files that are under 2 sec. Data/xxx/raw Transcription Install whisper pip install git+https://github.com/openai/whisper.git To turn off language auto-detection, set it to English only, and use model; edit as below: large short_audio_transcribe.py # set the spoken language to english
    print('language: en')
    lang = 'en'
    options = whisper.DecodingOptions(language='en')
    result = whisper.decode(model, mel, options)
	
    # set to use large model
    parser.add_argument("--whisper_size", default="large")

    #Solve error "Given groups=1, weight of size [1280, 128, 3], expected input[1, 80, 3000] to have 128 channels, but got 80 channels instead" while using large model
    mel = whisper.log_mel_spectrogram(audio,n_mels = 128).to(model.device) Run to start the transcription. python3 short_audio_transcribe.py Re-sample the sliced dataset: python3 resample.py --sr 44100 --in_dir ./Data/zizek/raw/ --out_dir ./Data/zizek/wavs/ Preprocess transcription: python3 preprocess_text.py --transcription-path ./Data/zizek/esd.list Generate BERT feature config: python3 bert_gen.py --config-path ./Data/zizek/configs/config.json Training and Inference Run to start training python3 train_ms.py Edit for model path: config.yml model: "models/G_20900.pth" Run to start webui for inference python3 webui.py vits-simple-api is a web frontend for using trained models. I use this mainly for its long text support which the original project doesn’t have. vits-simple-api git clone https://github.com/Artrajz/vits-simple-api
git pull https://github.com/Artrajz/vits-simple-api
cd vits-simple-api

conda create -n vits-simple-api python=3.10 pip
conda activate vits-simple-api && 

pip install -r requirements.txt (Optional) Copy pre-trained model files from to Bert-vits2-V2.3/ vits-simple-api/bert_vits2/ Copy and to Bert-vits2-V2.3/Data/xxx/models/G_xxxxx.pth Bert-vits2-V2.3/Data/xxx/config.json vits-simple-api/Model/xxx/ Edit for and as preferred config.py MODEL_LIST Default parameter Edit as below: Model/xxx/config.json "data": {
    "training_files": "Data/train.list",
    "validation_files": "Data/val.list",
	
  "version": "2.3" Check/Edit in as model_list config.yml [xxx/G_xxxxx.pth, xxx/config.json] Run python app.py Tweaks for tone, for randomness, for pronunciation, and for speed. and are self-explanatory SDP Ratio Noise Noise_W Length emotion style Share Models In its , there are a lot of VITS models shared by others. You can try it out first, and then download the desired models from . Hugging Face repo Files The is widely used in some content creation communities because of its high quality. It contains hundreds of characters, although only Chinese and Japanese are supported. Genshin model In , there are a lot of Bert-vits2 models that are made from popular Chinese streamers and VTubers. another repo There are already projects making AI Vtubers like and . I’m looking forward to how this technology can change the industry in the near future. this this https://techshinobi.org/posts/voice-vits/https://techshinobi.org/posts/voice-vits/

How to Voice Clone With SoftVC VITS and Bert-VITS2: A Deep Dive

SoftVC VITS এবং Bert-VITS2 দিয়ে কীভাবে ভয়েস ক্লোন করবেন: একটি গভীর ডুব

So klonen Sie Stimmen mit SoftVC VITS und Bert-VITS2: Ein tiefer Einblick

Cómo clonar por voz con SoftVC VITS y Bert-VITS2: una inmersión profunda

Comment cloner la voix avec SoftVC VITS et Bert-VITS2 : une plongée approfondie

SoftVC VITS और Bert-VITS2 के साथ वॉयस क्लोन कैसे करें: एक गहन जानकारी

SoftVC VITS および Bert-VITS2 を使用して音声クローンを作成する方法: 詳細

SoftVC VITS 및 Bert-VITS2를 사용한 음성 복제 방법: 심층 분석

Como clonar voz com SoftVC VITS e Bert-VITS2: um mergulho profundo

Как озвучить клон с помощью SoftVC VITS и Bert-VITS2: глубокое погружение

SoftVC VITS ve Bert-VITS2 ile Ses Klonlama: Derinlemesine Bir İnceleme

Cách sao chép giọng nói với SoftVC VITS và Bert-VITS2: Tìm hiểu sâu

如何使用 SoftVC VITS 和 Bert-VITS2 进行语音克隆：深入探讨

147 Stories To Learn About Reading Books Online

The holiday season brings the joyous task of finding that perfect gift for loved ones, friends, and the hard-to-shop-for savvy entrepreneurs in our lives. These go-getters are often on the move, their minds buzzing with the next big idea or strategy, leaving little time to indulge in the pleasures of a well-considered present. The challenge is magnified when the entrepreneur is a frequent traveler whose nomadic lifestyle is both an asset and a hurdle in the art of gift-giving. Digital gifts are a , noting their convenience and the personalized experiences they can offer. They are particularly relevant for the traveling entrepreneur, whose office may vary from a coffee shop in Berlin to an airport lounge in Tokyo. For them, a gift seamlessly integrating into their digital toolkit holds more value than any tangible item. growing trend The following sections will explore a curated selection of five perfect Saas-based gifts. They promise to be a hit with the entrepreneurs in your life, bringing them joy and utility long after the holiday decorations are packed away. Music and Podcasts on the Go For the entrepreneurs whose work takes them across continents, the gift of Spotify Premium is like a universal passport to a world of music and podcasts. Imagine them, headphones in ear, finalizing a business deal in London while listening to the latest episode of "How I Built This" or unwinding with their favorite album after a successful client meeting in New York. Spotify offers an ad-free experience with high-quality audio and the ability to download tracks for offline listening—perfect for those moments when Wi-Fi is as elusive as a quiet moment in a bustling airport terminal. Gifting Spotify is straightforward and thoughtful. Simply purchase a Spotify gift card from a retailer like Amazon or Best Buy or directly from the . These cards come in various denominations, offering one, three, six, or twelve months of Premium subscription. The recipient can redeem the card and enjoy uninterrupted music and podcasts on the fly. Prices range from $9.99 for a one-month subscription to $99 for an annual pass, making this an adaptable gift for any budget. Spotify website The ease of gifting and the rich content library make Spotify Premium an excellent choice for traveling entrepreneurs looking to stay inspired, informed, and entertained. Continuous Learning and Skill Enhancement In a world where business evolves at the speed of light, continuous learning is not just an advantage, it's a necessity. Entrepreneurs who travel frequently encounter diverse cultures, markets, and innovations that challenge their existing knowledge and skill set. MasterClass offers a unique opportunity for these individuals to learn from the very best in various fields, whether it's business strategy from Bob Iger or creative writing from Margaret Atwood. is as easy as visiting their website. You can select an annual membership, which provides unlimited access to all classes across various categories. This gift can be the key to unlocking new skills, gaining insights, or even finding a new passion. Gifting MasterClass An annual MasterClass membership costs around $180 and grants access to all courses for one year. Sometimes, during special promotions or the holiday season, MasterClass offers a buy-one-gift-one deal, making it even more appealing for those looking to treat themselves as well as gift an entrepreneur in their life. For the entrepreneur who values growth, inspiration, and learning from the best, a MasterClass subscription is a gift that keeps giving long after the holiday lights dim. It's an investment not only in their business but also in their personal development and lifelong learning journey. A Winning AI-Powered Virtual Team The modern entrepreneur is a master of multitasking, juggling multiple projects, and staying on top of the latest trends—all while traversing the globe. In this non-stop whirlwind of activity, a team of AI-powered virtual assistants can be the difference between chaos and order. Enter Olympia, the virtual team that harnesses the power of artificial intelligence to streamline an entrepreneur's busy life. Olympia delivers an advanced suite of AI-driven consulting services perfectly aligned with the dynamic lifestyle of the traveling businessperson. Equipped with a remarkable long-term memory, the Olympia AI team will learn every detail about your entrepreneurial friend's business, ensuring tailored support and seamless collaboration. It operates with a depth of personalization and efficiency within various domains like marketing, copywriting, business consulting, SEO, PR, legal, sales, communications,  pair programming, and more. Its ability to generate and edit downloadable long-form content, send emails, perform Google and web searches, and analyze data from uploaded documents without rate limits makes it an indispensable asset. Third-party integrations like HubSpot, soon-to-launch Google Suite, and more make it a real deal for any entrepreneur. This holiday season, enhance their business operations with the gift of Olympia's Pro Plus annual plan, now available at an exclusive 20% off as a , which makes it $599. By gifting an Olympia subscription, you're not just offering a tool; you're giving the gift of time—a resource more precious than any other for the busy entrepreneur. It's the kind of present that underscores your support for their ambitions and your belief in their potential, making it a truly meaningful offering for the holiday season or any time of year. Christmas Special Mental Health Support During Hectic Schedules A relentless pace and a never-ending to-do list often characterize the life of a traveling entrepreneur. While chasing business goals and hopping from one time zone to another, mental health can take a backseat. However, maintaining a healthy mind cannot be overstated, as it is the bedrock upon which creativity and productivity are built. This is where Calm, the app for meditation and sleep, comes into play. Calm offers a sanctuary amidst the storm of entrepreneurial responsibilities. With guided meditations, Sleep Stories, breathing programs, and relaxing music, the app is designed to reduce anxiety, improve sleep, and help users find moments of peace. For entrepreneurs battling stress and burnout, Calm provides the tools to develop a mindfulness practice that fits their hectic schedules. Gifting Calm to the busy entrepreneur is a thoughtful way of saying you care about their well-being. To gift a subscription, visit the Calm website's , where you can choose from several options, including 1-month, 1-year, and lifetime subscriptions. The cost of a Calm subscription varies, with a one-year subscription typically priced at $69.99, which offers complete access to all the app's features. Once purchased, you'll receive a gift code to present to the recipient, who can redeem it for full access to Calm's premium content. gift page Books for the Busy Traveler Reading, a vital avenue for learning and relaxation, can be challenging to fit into a busy schedule, particularly for those always on the road. This is where Audible shines as an ideal companion for the traveling businessperson. As the provider of audiobooks and spoken-word entertainment, Audible allows users to absorb the wisdom of books while in transit, whether navigating through airport terminals or en route to their next destination. Gifting an Audible subscription is a gesture that acknowledges the recipient's need for knowledge and personal development. To gift Audible, visit and select from 1, 3, 6, or 12-month membership plans. Each gift membership comes with a set number of Audible credits, which the recipient can use to purchase audiobooks. this page An Audible gift membership starts at $15.00 for a 1-month subscription, with the price increasing accordingly for longer durations. The 12-month option, priced at $150.00, offers the best value with 12 credits and access to the Audible Plus catalog, providing an entire year of auditory enlightenment. For the entrepreneur who constantly seeks to maximize every moment, an Audible subscription is a gift that will be appreciated deeply. It's a practical, thoughtful present that fits seamlessly into their dynamic lifestyle, providing inspiration, education, and entertainment wherever their business takes them. Also appears . here

5 Wonderful SaaS-tastic Gift Ideas for Savvy Entrepreneurs

240 Stories To Learn About Startups Of The Year

Dedicated to Waterstone's, the national UK bookselling chain

Little Brother by Cory Doctorow, is part of the HackerNoon Books Series. You can jump to any chapter in this book . OTHER EDITIONS here OTHER EDITIONS My agent, Russell Galen (and his sub-agent Danny Baror) did an amazing job of pre-selling rights to Little Brother in many languages and formats. Here's the list as of today (May 4, 2008). I'll be updating it as more editions are sold, so feel free to grab another copy of this file (http://craphound.com/littlebrother/download) if there's an edition you're hoping to see, or see http://craphound.com/littlebrother/buy/ for links to buy all the currently shipping editions. Audiobook from Random House. A condition of my deal with Random House is that they're not allowed to release this on services that use "DRM" (Digital Rights Management) systems intended to control use and copying. That means that you won't find this book on Audible or iTunes, because Audible refuses to sell books without DRM (even if the author and publisher don't want DRM), and iTunes only carries Audible audiobooks. However, you can buy the MP3 file direct from RandomHouse or many other fine etailers, or through this link My foreign rights agent, Danny Baror, has presold a number of foreign editions: Greece: Pataki Russia: AST Publishing France: Universe Poche Norway: Det Norske Samlaget No publication dates yet for these, but I'll keep updating this file as more information is available. You can also subscribe to my mailing list for more info. About HackerNoon Book Series: We bring you the most important technical, scientific, and insightful public domain books. This book is part of the public domain. Cory Doctorow (2009). Little Brother. Urbana, Illinois: Project Gutenberg. Retrieved https://www.gutenberg.org/cache/epub/30142/pg30142-images.html This eBook is for the use of anyone anywhere at no cost and with almost no restrictions whatsoever. You may copy it, give it away or re-use it under the terms of the Project Gutenberg License included with this eBook or online at , located at . www.gutenberg.org https://www.gutenberg.org/policy/license.html

OTHER EDITIONS

Natural Language Processing (NLP) has emerged as a transformative force that reshapes how we interact with information and communicate with machines around the world. NLP is a field at the intersection of computer science, linguistics, and artificial intelligence, focusing on enabling computers to understand, interpret, and generate human language in a way that mirrors human cognition. As NLP continues to advance in various industries such as healthcare, finance, customer service, and others around the world, it is important to gain practical experience through hands-on NLP projects to become a good Data scientist or NLP engineer. The scope of NLP applications is both vast and diverse, covering a wide range of industries and use cases. From sentiment analysis and chatbots to language translation, speech recognition, and information retrieval. NLP-driven applications enhance search engines' accuracy, automate customer interactions, facilitate multilingual communication, and even assist in legal document analysis. This variety highlights how flexible NLP is and emphasizes the urgent requirement for experts who are knowledgeable in its techniques. Textbook knowledge and theoretical understanding are still valuable components of learning NLP, but they can only take an individual so far. The true mastery of NLP comes from hands-on experience, where learners engage themselves in real-world projects to tackle challenges, experiment with various algorithms, and overcome practical challenges. This will help you to gain invaluable insights about preprocessing text data, feature engineering, selecting appropriate models, fine-tuning parameters, and effectively evaluating results. In this article, you will learn different NLP project ideas that focus on practical implementation to help you master the NLP techniques and be able to solve different challenges. Sentiment Analysis Sentiment analysis is an NLP technique that involves determining the sentiment or emotional tone behind a piece of text, such as a review, tweet, or customer feedback. The main goal of sentiment analysis is to categorize the sentiment expressed in the text as positive, negative, or neutral. This process is important for understanding public opinion, making informed business decisions, monitoring brand reputation, and evaluating customer satisfaction. In the era of social media and online reviews, sentiment analysis helps businesses understand customer feedback at scale, enabling them to identify areas of improvement and improve customer experience. It assists in monitoring and managing brand perception, as well as predicting market trends based on sentiment shifts. Several datasets are available for training and evaluating sentiment analysis models. These datasets are often labeled with sentiment labels (positive, negative, neutral) to facilitate supervised machine learning. Some popular datasets include: A dataset containing movie reviews with binary sentiment labels (positive/negative). It is widely used for sentiment analysis model benchmarking. IMDb Movie Reviews: This dataset contains reviews of various products sold on Amazon, and the reviews are annotated with sentiment labels. Amazon Product Reviews: Datasets of tweets labeled with sentiment labels, commonly used for social media sentiment analysis. Twitter Sentiment Analysis: Creating a sentiment analysis project involves a combination of programming languages, libraries, and tools. The tech stack includes Python, a popular langauge in NLP for its libraries; NLTK for various NLP tasks; Scikit-Learn for machine learning; TensorFlow or PyTorch for deep learning; Pandas for data manipulation; SQLite or MySQL for efficient data storage; and GitHub or GitLab for version control and collaboration with others. Here is a list of few NLP projects on sentiment analysis you can start with: Ecommerce product reviews - Pairwise ranking and sentiment analysis Many-to-One LSTM for Sentiment Analysis and Text Generation Text Classification Text classification in NLP involves the process of automatically categorizing or labeling pieces of text into predefined categories or classes based on their content and meaning. This task is aimed at teaching computers to understand and organize large amounts of text data, such as emails, articles, or social media posts, by assigning them to specific categories like spam or not spam, topics like sports or technology, and more. Text classification serves as a cornerstone in information organization by enabling the systematic categorization of textual content. This categorization enables businesses, researchers, and individuals to quickly access, sort, and analyze information. There are several publicly available datasets that cover a wide range of text classification tasks, such as spam detection, topic classification, and more. Examples include the 20 Newsgroups dataset for topic classification and the Enron email dataset for email categorization. Creating a text classification project involves assembling a suitable tech stack that leverages the power of NLP libraries and machine learning frameworks. You can use NLTK for various NLP tasks; Scikit-Learn for machine learning; TensorFlow or PyTorch for deep learning; Pandas for data manipulation; SQLite or MySQL for efficient data storage; and GitHub or GitLab for version control and collaboration with others. Here is a list of few NLP projects on text classification you can start with: Build a Multi ClassText Classification Model using Naive Bayes PyTorch Project to Build a LSTM Text Classification Model Topic Modeling Topic modeling in NLP is a technique that involves automatically identifying and extracting the main themes or topics present in a collection of texts. It aims to uncover the underlying structure within the text data by grouping together words that frequently appear together and represent coherent subjects. This helps in gaining insights into the main subjects discussed in the documents and enables various applications like content recommendation, information retrieval, and summarization. Datasets for topic modeling in NLP include various text sources like news articles, academic papers, social media posts, reviews, blogs, legal documents, and more. These datasets are used to automatically identify and extract main subjects within texts. Depending on the application, datasets may include healthcare records, email archives, or specialized domain-specific data. To create a topic modeling project, a tech stack may involve Python programming language and libraries like NLTK or spaCy for text processing, Scikit-Learn for machine learning tasks, and Gensim for topic modeling algorithms. Deep learning frameworks like TensorFlow or PyTorch can be used for advanced topic modeling approaches, with Pandas for data manipulation and SQLite or MySQL for efficient data storage. Version control is managed through platforms like GitHub or GitLab, combination of tools that covers the project's various stages, from data preprocessing to model training and predictions. Here is a list of few NLP projects on Topic modeling you can start with: Topic modeling using Kmeans clustering to group customer reviews NLP Project on LDA Topic Modelling Python using RACE Dataset Name Entity Recognition Named Entity Recognition (NER) is an NLP task that involves identifying and classifying specific entities, such as names of people, places, organizations, dates, and more, within text. NER aims to automatically categorize these entities to provide structure and meaning to unstructured text data, enabling information extraction, content analysis, and information retrieval. NER is used across different fields such as information retrieval,chatbots, financial analysis, healthcare, and news categorization to automatically identify and classify specific entities in text which helps to improve search & content analysis and decision-making in various industries. Datasets tailored for NER tasks contain text with annotated instances of named entities and their corresponding categories, serving as training and evaluation material for NER models. Common datasets include CoNLL-2003 for English NER, Groningen Meaning Bank (GMB) for English entities, and masakhaNER for african languages. Creating a Named Entity Recognition (NER) project involves assembling a tech stack that includes Python, NLP tools like spaCy or NLTK for entity recognition, machine learning frameworks such as Scikit-Learn for feature engineering, and deep learning platforms like TensorFlow or PyTorch for neural network-based models. NER-specific libraries like Flair or AllenNLP enhance the process, Together, this stack facilitates the comprehensive development of NER models. Here is a list of few NLP projects on NER you can start with: NLP Project to Build a Resume Parser in Python using Spacy MasakhaNER: Named Entity Recognition for African Languages Machine Translation Machine translation in NLP refers to the automated process of translating text or speech from one language to another using computational techniques and algorithms. This process involves teaching computers to understand the meaning and structure of a source language text(e.g English) and generate an equivalent text in a target language (e. Swahili). In tourism and travel, machine translation helps people who visit different places and speak different languages. It translates things like menus, signs, and travel guides, making it easier for travelers. In government diplomacy, machine translation helps countries talk to each other by translating important papers and messages. This helps countries work together and understand each other better. Machine translation datasets contain sentence pairs in different languages to train and test translation models. These datasets include parallel corpora like Europarl and MultiUN user-contributed translations, and more. Custom datasets can be created for specific domains. The machine translation tech stack includes programming languages like Python, NLP libraries such as spaCy, specialized machine translation frameworks like OpenNMT, pre-trained models like Transformers, data processing tools like Pandas, and alignment/tokenization tools. Deep learning frameworks like TensorFlow or PyTorch are used for training the models. The stack enables the development of translation models, covering data preprocessing, model training, deployment, and evaluation. Here is a list of few NLP projects on machine translation you can start with: A Machine Translation project that translates text from English to French English to Italian Neural Machine Translator Question Answering Question Answering (QA) in NLP refers to the automated process of extracting precise answers from a given text or document in response to user-generated questions. QA systems aim to understand the meaning of the questions and the context of the text to locate relevant information and generate accurate answers. These systems can be applied to various domains, such as search engines, customer support, educational platforms, and information retrieval, enabling users to quickly obtain specific information without manually reading through extensive texts. Datasets used for Question Answering tasks contain pairs of questions and corresponding answers and come in various formats and types to cover different types of questions and texts. Some common types of QA datasets include: A widely used dataset with questions sourced from Wikipedia articles and their corresponding paragraphs containing answers. SQuAD (Stanford Question Answering Dataset): A dataset containing questions from trivia competitions, sourced from Wikipedia, and including evidence documents. TriviaQA: Questions created by humans based on news articles, with corresponding sentences serving as answers. NewsQA: Creating a QA project involves a specific tech stack that includes using programming languages like Python, and libraries such as spaCy or NLTK for text preprocessing and linguistic analysis. Deep learning frameworks like TensorFlow or PyTorch are utilized for building and training QA models. Specialized QA libraries like Hugging Face's Transformers provide pre-trained models and tools for QA tasks. Here is a list of few NLP projects on Question Answering you can start with: Question Answering On SQUAD FinBERT-QA-notebooks Automatic Speech Recognition Automatic Speech Recognition (ASR) in NLP refers to the technology that converts spoken language into written text. ASR involves the use of computational algorithms and models to transcribe spoken words from audio recordings or real-time speech into accurate and readable text format. ASR has a wide range of applications, including transcription services (Ref), voice assistants (Amazon Alexa, Apple Siri, and Google Assistant) and more, enabling human-computer interaction through spoken language. Datasets used for developing Automatic Speech Recognition (ASR) systems consist of paired audio recordings and their corresponding transcriptions in text format. These datasets are crucial for training and evaluating ASR models. Some commonly used ASR datasets include: An open-source dataset with multilingual audio recordings and transcriptions contributed by volunteers, used to build ASR models for various languages around the world. CommonVoice: This dataset contains audiobooks with aligned transcriptions, providing a diverse range of speech patterns and accents. LibriSpeech: Organizations or communities can create their own datasets by recording speech related to specific domains or industries. Custom Created Datasets: Creating an Automatic Speech Recognition (ASR) project involves a tech stack including programming languages like Python, audio processing libraries such as librosa, specialized ASR libraries like Kaldi or Mozilla DeepSpeech, Nemo,  deep learning frameworks like TensorFlow or PyTorch for model development, ASR-specific libraries like SpeechRecognition or vosk for integration, and data augmentation tools like SoX for enhancing the dataset. Here is a list of few NLP projects on speech recognition you can start with: End-to-End Speech Emotion Recognition Project using ANN Speech Recognition Macedonian Conclusion NLP project-based learning offers hands-on experience, allowing you to apply theoretical knowledge to real-world situations. This approach promotes critical thinking, problem-solving, and creativity while encouraging collaboration and teamwork. Engaging in projects helps you to gain practical skills in coding, data manipulation, model building, and deployment. Also, it helps to improve your employability and confidence. It's important to select projects that resonate with your passions and align with your expertise. Choosing projects that genuinely interest you keeps motivation high and makes the learning experience more enjoyable. Leveraging your existing skills and knowledge ensures a smoother learning curve and a higher chance of success. By aligning projects with your interests and expertise, you'll not only maximize your learning but also create valuable outcomes that reflect your strengths and dedication.

7 NLP Project Ideas to Enhance Your NLP Skills

আপনার এনএলপি দক্ষতা বাড়ানোর জন্য 7টি NLP প্রকল্পের ধারণা

7 NLP-Projektideen zur Verbesserung Ihrer NLP-Fähigkeiten

7 ideas de proyectos de PNL para mejorar tus habilidades de PNL

7 idées de projets PNL pour améliorer vos compétences en PNL

आपके एनएलपी कौशल को बढ़ाने के लिए 7 एनएलपी परियोजना विचार

NLP スキルを向上させる 7 つの NLP プロジェクトのアイデア

NLP 기술을 향상시키는 7가지 NLP 프로젝트 아이디어

7 ideias de projetos de PNL para aprimorar suas habilidades de PNL

NLP Becerilerinizi Geliştirecek 7 NLP Proje Fikri

7 ý tưởng dự án NLP để nâng cao kỹ năng NLP của bạn

提高 NLP 技能的 7 个 NLP 项目想法

Realistic Text-to-Speech Voice Synthesis: Comparing Tortoise and Bark

In the realm of text-to-audio AI models, the race to create more realistic, immersive experiences has been heating up. A promising runner in this race is AudioLDM, which uses some novel techniques to generate speech, sound effects, or music from text or audio prompts. But how exactly does AudioLDM achieve this? And more importantly, what does the way in which the tech works tell us about the kinds of products we can build with it? In this article, we'll study AudioLDM to see what makes it special and understand how we can take advantage of those unusual and special capabilities to build interesting new startup products. Let's begin. Subscribe or follow me on Twitter for more content like this! AudioLDM at a glance LDM stands for Latent Diffusion Model. AudioLDM is a novel AI system that uses latent diffusion to generate high-quality speech, sound effects, and music from text prompts. It can either create sounds from just text or use text prompts to guide the manipulation of a supplied audio file. Check out from the 's project website, showing off some of the unique generations the model can create. I liked the upbeat pop music the best, and I also thought the children singing, while creepy, was pretty interesting. this video demo creator Key features of AudioLDM: Uses a Latent Diffusion Model (LDM) to synthesize sound Trained in an unsupervised manner on large unlabeled audio datasets Operates in a continuous latent space rather than discrete tokens Employs Cross-Modal Latent Alignment Pretraining (CLAP) to map text and audio Can generate speech, music, and sound effects from text prompts a combination of a text and an audio prompt or Allows control over attributes like speaker identity, accent, etc. Synthesizes realistic sound effects and music Creates sounds not limited to human speech (e.g. nature sounds) Generates audio samples up to 10 seconds long In a nutshell, AudioLDM pushes boundaries in controllable audio synthesis and lets us build some innovative text-to-audio products and services that were previously difficult to build. There are opportunities in gaming, audiobooks, accessibility, and beyond. Let's see a few. AudioLDM Use Cases There are lots of world-changing AI products just waiting to be built, all powered by AudioLDM. The capability to create high-quality, customizable synthetic sound isn't just an interesting theoretical outcome to read about on Arxiv—it also enables a host of new impactful applications for text-to-audio technology. Here are a few: : With AudioLDM, you can break the shackles of generic gaming dialogues. Imagine games where every character has a unique voice and speech pattern. Your games will no longer be about just play—they become living, breathing digital realms filled with distinct personalities that heighten immersion for gamers. You can also use the model to create sound effects from just a text prompt. Unleash Immersive Gaming Experiences : It's time to take the listener on a journey. Use AudioLDM to create audiobooks where each character speaks in a unique voice. Transform the listening experience from a monologue to a dynamic conversation, and create theatre-like immersion for your audience. Revolutionize the Audiobook Landscape : Learning is personal, and so should be the learning platforms. Use AudioLDM to offer personalized voices that cater to individual learner needs in your language learning apps or online media. Make the learning process more efficient, enjoyable, and inclusive. Reimagine Learning Platforms : With AudioLDM, you can make technology more accessible for individuals with disabilities. Whether it's screen readers or voice assistants, you can customize and control the synthesized speech to make your user experience more personal and effective. Drive Accessibility Forward As you can see here, there are a lot of potential startups waiting to be built. How are you going to use AudioLDM to build a text-to-audio product that will reshape your industry? Let's peel back the layers to understand what makes AudioLDM tick, and how you can harness its power to fuel your startup's growth. The Technology Behind AudioLDM If you're considering AudioLDM as a tool to create your next product, you need to understand the distinct technological principles that make this AI model unique. AudioLDM's text-to-audio capabilities are built on three pillars: its usage of 'unpaired learning', its operation within a 'continuous latent space', and a special technique known as 'Cross-Modal Latent Alignment Pretraining' (CLAP). Each of these elements contributes to its ability to synthesize controllable, high-quality sound. Let's look at each element in more detail to understand why they matter. Unpaired Learning Most AI models expect tidy datasets of aligned input-output pairs during training - for example, large tables of audio clips perfectly matched to their text transcriptions. AudioLDM rips up this playbook, learning entirely from audio data without any accompanying text. unpaired It relies on a clever self-supervised pretraining scheme. The model tries to reconstruct audio waveforms from compressed latent representations of those sounds. By minimizing the reconstruction error, it's forced to encode all the most salient features of natural speech into the latent space. Once AudioLDM has learned to represent audio data in this compact form, it can then condition its internal generative model on text prompts. This unpaired approach means we bypass cumbersome labeled datasets, instead unleashing the model on oceans of unlabeled audio data. Latent Space vs. Tokens Many speech models generate audio as a series of discrete, categorical outputs - like putting together Lego blocks to form sounds. But AudioLDM breaks free of this restrictive paradigm. It operates in a smooth, continuous latent space obtained by condensing audio spectrograms using a variational autoencoder (VAE). A spectrogram is a visual representation of the spectrum of frequencies in a sound over time. It displays the intensity levels of different frequencies as they vary through an audio clip. Spectrograms are created by applying a Fourier transform to break down sound waves into component frequencies. A is a type of neural network used for dimensionality reduction and generative modeling. It consists of an encoder and a decoder. The encoder compresses input data into a lower-dimensional latent space. The decoder tries to reconstruct the original input from points in the latent space. VAEs are used in many different models, and are often used in image generators like . variational autoencoder (VAE) Anything v3 VAE In a VAE, the encoder outputs a mean and variance for each latent dimension. These define a Gaussian distribution in the latent space. Sampling from these distributions and passing samples through the decoder allows generating new data similar to inputs. So for AudioLDM, the VAE condenses audio spectrograms into a smooth latent space. Interpolating between latent vectors mixes different voices and sounds. The decoder converts latent points back into realistic audio by leveraging patterns learned from training data. This gives AudioLDM its expressive generative capabilities. This continuity also encodes natural variation and uncertainty inherent to speech. Sampling latent points results in diverse outputs, even when the text prompt stays fixed. And lower-dimensional latent modeling greatly reduces computational overhead. Bridging Modalities with Cross-Modal Conditioning Here's the trick that lets AudioLDM synthesize speech without explicitly training on aligned text-audio pairs. It uses CLAP (Cross-Modal Latent Alignment Pretraining) to create a shared embedding space between both modalities. Essentially, CLAP teaches the model the intrinsic relationships between textual concepts and their associated acoustic patterns. AudioLDM taps into these co-embeddings, using the audio-side to represent its internal generative model and the text-side to provide conditional inputs that guide the speech output. This technique gives AudioLDM huge flexibility in synthesizing voices from free-form text prompts. Even for unusual text inputs, the cross-modal embeddings convey informative guidance rooted in the structures of natural speech. In the paper, for example, the creators demonstrate that AudioLDM can create sounds from very unusual text prompts, like "a wolf is singing a beautiful song." So by breaking conventions with unpaired learning, opening up greater possibilities through continuous latent modeling, and leveraging cross-modal conditioning, AudioLDM manages to unlock new horizons for controllable speech synthesis. Its technical inventiveness has opened exciting doors for creating more immersive, expressive conversational experiences. AudioLDM's strengths and weaknesses While AudioLDM demonstrates leading-edge achievements in controllable audio synthesis, it excels better at some tasks than others. Let's examine its capabilities and limitations for critical audio generation use cases. Diverse Audio Creation and Speech Synthesis AudioLDM shines at producing novel sound effects and music described in text prompts. Training on diverse datasets like AudioSet equips it to synthesize a wide range of sounds. Its continuous latent space enables smooth interpolations for generating musical notes and transitions. These traits make AudioLDM a versatile tool for audio creators. Intelligible but Imperfect Speech For speech synthesis, AudioLDM can generate intelligible voices from text and manipulate attributes like speaker identity based on prompts. It's capable of generating intelligible speech from text and can use attributes like the speaker's identity based on prompts. However, speech quality falls short of specialized . Subtleties like emphasis and cadence may not translate accurately, and artifacts are often audible. state-of-the-art TTS models Long-Form Audio Still a Challenge A current limitation is that AudioLDM works best for short clips of up to 10 seconds. Maintaining consistency over longer timescales is difficult as outputs are produced in small steps. Generating coherent long-form audio like podcasts or audiobooks remains an open research problem. You may wish to check into other models for these kinds of applications. auto-regressively The Nuances of Speech and Data Dependencies Despite its groundbreaking achievements, AudioLDM struggles with the complexity of human speech. While it can generate a diverse array of voices, it often fails to replicate the subtle emotional nuances and natural cadence inherent to human speech. This means elements like emphasis, sarcasm, and vocal richness are sometimes lost in translation. Further, AudioLDM's unpaired training scheme, although reducing the need for labeled data, remains a data-hungry beast. The model requires massive quantities of unlabeled audio data for pretraining, which may be beyond the reach of early-stage startups. Data partnerships could offer a solution, but come with their own set of legal and privacy issues. The complexities of AudioLDM extend beyond data needs. The model's advanced neural architecture demands significant computational resources for training, which can pose a challenge for startups with limited hardware capabilities. While cloud-based AI services could offset these demands, they would inevitably incur considerable costs. Precision Editing Limitations As a generative model, AudioLDM is designed for stochastic synthesis rather than precise audio editing. Operations like voice conversion or sound filtering require specialized techniques beyond their capabilities. If editing precision is needed, AudioLDM may not suffice in its current form. Black Box Outputs Like many neural networks, the internal workings of AudioLDM are opaque. This makes debugging quality issues and diagnosing unintended biases in the synthesized voices more difficult. Mechanisms to peek inside the black box and gain insight into failure modes would aid responsible development. By knowing about these limitations, you can make informed decisions about integrating AudioLDM into products, and direct research efforts towards mitigating the vulnerabilities. AI advances like AudioLDM inevitably bring new challenges alongside their benefits. Keeping grounded about their shortcomings is crucial to steer progress in the right direction. Comparing Apples to Apples: AudioLDM Vs. the Rest There are several AI models for generating audio from text prompts. Here we'll compare some of the popular options to AudioLDM to help determine the right fit for your needs. Riffusion uses stable diffusion to generate musical audio in real-time. It's useful for building things like interactive music apps. AudioLDM can also create music, but isn't specialized for real-time generation. Consider Riffusion if you need dynamic, improvisational music. Riffusion Musicgen converts text prompts into musical compositions. This makes it useful for assisting musicians, composers, or music students. AudioLDM has wider sound synthesis capabilities beyond just music. Musicgen may offer higher quality and more control for music specifically. Musicgen Bark generates realistic speech and sound effects from the text. The speech sounds more natural than AudioLDM and Bark is great to use if you're building voice assistants or audiobooks. The bark is focused on high-fidelity speech, while AudioLDM has a greater diversity of sound effects. Bark may be preferable for voice-centric applications. Bark Tortoise TTS converts text to natural-sounding speech. Like Bark, it's useful for voice assistants, audiobooks, and speech synthesis. Tortoise TTS specializes in human-like voices, but AudioLDM enables greater voice control and variation. Choose based on speech quality needs. Tortoise TTS Summary For specialized music generation, consider Riffusion or Musicgen. For voice-centric applications, Bark or Tortoise TTS may be ideal. AudioLDM provides versatile sound synthesis and voice control, but less focus on tailored domains. Here is a comparison table for the audio generation models, using data from . AIModels.fyi Model Creator Cost per Run Avg Run Time Hardware Use Cases Riffusion riffusion $0.0066 12 seconds Nvidia T4 GPU Real-time music generation, interactive music apps Musicgen joehoover $0.0943 41 seconds Nvidia A100 GPU Assisting musicians, composers, music students Bark suno-ai $0.0297 54 seconds Nvidia T4 GPU Voice assistants, audio books, sound effects Tortoise TTS afiaka87 - - Nvidia T4 GPU Voice assistants, audio books, speech synthesis Wrapping Up: Is AudioLDM the best model for your text-to-audio product? AudioLDM demonstrates cutting-edge innovation in controllable audio synthesis. Its novel techniques open up new creative possibilities that were once inconceivable. However, as with any new technology, there are tradeoffs to weigh before integrating it into products. Consider these key questions: Models like Riffusion, Musicgen, Bark, and Tortoise TTS may be better suited. Does your application require specialized music generation or exceptionally natural speech? If not, opt for lighter and more accessible alternatives. Can you accommodate AudioLDM's heavy data and compute requirements? AudioLDM is better for generative use cases than editing workflows. Is precision editing critical? AudioLDM's "black box" nature may pose challenges. Are model interpretability and transparency important? If not, you may be able to leverage a model hosting provider or platform like Replicate, but at an increased cost. Does your team have ML ops maturity to responsibly develop with AudioLDM? More specialized models could be a better fit. Are your end goals focused on niche audio domains? By weighing these key factors against the breakthrough capabilities AudioLDM provides, you can determine if now is the right time to pilot it in your products. With diligent evaluation, you can strategically harness AudioLDM's power to provide delightful user experiences and disrupt industries. Remember that AI models are improving all the time, and future advances may change key factors that could tip the scales in favor of implementing this model. Be sure to keep a like this one to see how they improve and change. Thanks for reading! close eye on models Subscribe or follow me on Twitter for more content like this! Resources and Further Reading You might find some of these other articles interesting or helpful in your quest to build your AI project. - Overview of the Whisper speech-to-text model. Converting Speech into Text with OpenAI's Whisper Model - Explains how AudioLDM synthesizes speech and sound from text. Audio LDM: AI Text to Audio Generation with Latent Diffusion Models - Guide to making your friends lip sync with AI models. Learn How to Harness the Power of AI for Lip Syncing Videos - Compares image upscaling models. Another technical dive similar to this one. Real-ESRGAN vs. Real-ESRGAN: From Theoretical to Real-World Super Resolution with AI Also published . here

A Creator's Guide to Synthesizing Sound Effects, Music, and Dialog With AudioLDM

Let's learn about via these 44 free stories. They are ordered by most time reading created on HackerNoon. Visit the to find the most read stories about any technology. Bots /Learn Repo There's no such thing as bad bots. It's just people using bots to do bad things. 1. How Did We Bypass the Sweeper Bot and Saved $26K Tokens in the Dark Forest of Blockchain ? How did we bypass the sweeper bot and saved $26K tokens in the Dark Forest of Blockchain ? 2. Top Chatbot Trends That Will Dominate Businesses in 2023 Chatbots are a type of artificial intelligence that is programmed to simulate human communication. They are able to understand the context of a conversation and provide answers in a natural-sounding way. 3. Politics in Social Media: Bots, IDM and Decentralized Moderation The internet has become the agora of modern times. The question is not whether we are discussing politics over the internet or not (we obviously are) but if... 4. How to Create a Telegram Bot Using Ruby I was tasked with creating a telegram bot using ruby a while back. I had no idea where to start and just like any other programmer out there I turned to the internet for clues. I did not find a good article and hence experienced a lot of problems creating the bot. I have decided to write an article to explain how to create a telegram bot from scratch. 5. Artificial Intelligence: The Future Of Travel Industry Machines may not have taken over the world yet, but they are seeping into our lives and making it better. AI is changing every aspect of our lives. From self-driving cars to talking bots, there are so many examples of AI in use today. The technology is disrupting so many industries, and the travel industry isn’t exempt from this. Gone are the days when you had to count on a travel agent to plan your next vacation. You no longer need to engage yourself in that tiresome chain of conversations with your agent for your travel arrangements. AI now infiltrates every aspect of the travel industry. With the help of this technology, you can get personalized and intelligent travel solution tailored according to your needs. 6. Trading Bots What are Trading Bots in general? Are trading bots useful? 7. 4 Reasons Why Email Is Obsolete, and You Should Move On Chances are, your business’ primary mode of internal communication is email. And you’re using instant messaging (IM) apps for virtual meetups and updates from the team, thanks to the coronavirus. Don’t get me wrong, email is great and all for internal communication (if you lived in the 20th century that is). But did you know your email to your teammates gets lost in a sea of unimportant emails? And irrelevant emails account for about 62% of the total emails in an average inbox! 8. Top 17 Instagram Bots To Get More Followers It can be a real challenge getting your footing on Instagram. Over the last few years it’s become a highly competitive platform for marketers, brands, and influencers. 9. Creating Telegram Bots with n8n, a No-Code Platform I am a big fan of Telegram because of its great bot-ecosystem. I have been planning to make a bot for a while and have been looking for some nice workflow tools that can help me accomplish handling the back-end workflow of the bots visually. Recently, I started working with n8n, a fair-code licensed tool that helps you automate tasks, sync data between various sources and react to events all via a visual workflow editor. 10. An In-depth Guide to Depop Bots Here’s everything you need to know about depop bots. 11. Top 10 Questions Managers are asking in REAL Team Meetings to 🚀 productivity My team meeting is crucial. I get everyone in a room and get everything I need out of them — with the added bonus of the team getting in sync, right? Wrong. Turns out, managers are blind to how crucial that meeting is to others. Only 4–9% of problems are known to managers... gulp 12. Trading Bots vs Humans · Everything you need to know Over the past 10 years we've seen the rise and rise of trading bots and Quantitative Funds and we've seen the fall and fall of traditional Asset Managers and Hedge Funds. 13. How To Make A Chatbot From Scratch How To Make A Great Chatbot From Scratch Expert tips to create a highly functional chatbot. 14. 5 Chatbot Ideas Businesses Should Consider in 2019 If you are looking to add the most advanced chatbot to your website, you would have probably noticed that there are many things required to develop a chatbot. 15. How to Build a Bitcoin Telegram Bot I created a simple Telegram bot that could retrieve Bitcoin's Price to convince my friends that Telegram is far superior to Whatsapp. 16. Telegram Bots Have Helped Pave The Way for dApps 2020 is shaping up to be the year of dApps! New communities are forming around cryptocurrency and they are asking for just one thing: decentralization. All the services and platforms you are accessing online right now can be decentralized in one way or another. 17. How to Build Telegram Chats with a Crypto-trading Bot Perhaps a sign of the times: my most active Telegram chat is with a crypto-trading bot that constantly listens for opportunities to trade on my behalf. I used an open-source library to develop some strategies and configure the bot to execute them using my Binance account. The bot communicates all of its trades through Telegram and can reply to my requests to take action or share live updates. 18. Bot Strategies: Manipulating Democratic Discourse Two-thirds of Americans know what bots are, from a strictly negative perspective—and not without good reason. Malicious bot activity has evolved over time, from hackers using bots to infect millions of devices with malware to the more modern usage as spreaders of propaganda, fake news, and false social media accounts. 19. Build a WhatsApp Chatbot With Python, Flask, and Messagebird Tutorial, about building a chatbot using the Messagebird API for WhatsApp and the Flask framework for Python. 20. Let's Build a Telegram Bot in Rust. Part 1 Building telegram bot using Rust. 21. 7 AI-powered Chatbots If you’re a millennial, you’ll know SmarterChild, the first-ever instant messaging bot with natural language comprehension ability. It was developed in 2000 and demonstrated exceptional wit, which most of today’s bot cannot. SmarterChild used to chat with about 2,50,000 humans every day with funny, sad, and sarcastic emotions. Today, we’ve traveled a distance with technologies like AI, ML, NLP, etc. and bots like Xiaocle have passed Turing tests of 10 minutes (i.e. users couldn’t identify that they’re talking to a bot for about 10 minutes). 22. IEO and Trading Bots to Help You Win Each Battle Post by Tim Kozak, Head of Technology at Intellectsoft Blockchain Lab 23. 7 Reasons to Start Automating Business with Bots (Original: Unsplash) 24. Introduction to Chatbot Development Now-a- Days chatbots are playing an essential role in the day to day activities of humans {personals and organizations}. According to the survey done, 27% of the people are using the AI-powered chatbots such as Google Assistant, Amazon Alexa, Cortana, Apple Siri etc. for simple task or commands. 25. 5 Case Studies that Prove Bots Are Here to Help Businesses Scale It was about three years ago that Microsoft CEO, Satya Nadella, was quoted stating “Bots are the new apps,” during a 3-hour keynote to kick off the company’s Build conference. That statement has probably never been truer, especially since NLP bots Enterprise bots have appeared on the scene. 26. Wordpress Security Tutorial: How to Defend Against Bots With reCAPTCHA How To Increase Your Wordpress Security And Defend Your Applications From Automated Bots. Learn how to integrate reCAPTCHAs into Wordpress. 27. What Are The Different Types of  Malware? A walkthrough the different types of malware - from viruses to botnets, ransomware and phishing. Learn more and get protected! 28. Creating a Telegram Bot for my Friend's Business Last week my friend asked for help with coding a Telegram bot for the VIP 😏 customers of his herbs shop. 29. Saying No to Bots is Saying No to Automation Why You Should Start Using Bots Today 30. Spam Bots and Why They Matter in Elon Musk's Twitter Deal 31. Creating Sustainable Web3 Games: Existing Challenges and Potential Solutions Check the obstacles currently standing in the way of sustainability in Web3 gaming and the solutions that can drive the next wave of game development in GameFi 32. How To Convert PDFs Into AudioBooks It's about How To Convert PDFs Into AudioBooks With 2 Lines of Python Code. It'll make life easy for many lazy people . 33. 6 Popular Software Testing Trends Everyone Should Follow 34. Building Your Own Discord Bot With Python Is Easy Create your own Discord Server in just Four Lines of Code with Python 35. The 6 Best Platforms To Create Chatbots Without Additional Programming Today I want to share a small review of simple platforms for creating bots without any knowledge of programming languages 36. From Zero to 45k Users: How We Created a Chatbot To Track Airfares Everything started when my mate was going on a vacation to Spain. Every single day, he checked the airlines’ websites to find the cheapest ticket and manually monitored prices of a certain flight. All these attempts were tedious, humdrum, and time-consuming for him. 37. Crypto-Trading Bots: An Overview As the mainstream popularity of cryptocurrency trading grows, more and more people are looking for a way to casually get into it. Many realize the difficulties, as well as the risks of trading, thus trying to ease into the process without putting too much on the line right away. 38. CI/CD Isn't Just About Efficiency I’d like to talk about something we don’t bring up quite as often. Put simply, CI/CD makes us happy. 39. How to Add a Voice Assistant to your Mobile App? Don’t you think that a great many mobile apps would be a lot more convenient if they had voice control? And I don’t mean chatting with a banking bot. In most cases, voice navigation or a conversational form-filling is just enough. 40. Enterprise Chatbot: Do You Really Need It? This Will Help You Decide 41. Data Privacy In Social Media Matters and Here's Why In the past few years, privacy fears amongst social media users have augmented tremendously. Many users are reconsidering their interactions on social media and protecting their personal information due to the ever-increasing data infringement cases. 42. Journey to the Center of the Unseen World of Discord Bots If you've used Discord for any amount of time, I'm sure that you've come across at least a few Discord bots. Given MEE6 or Groovy, they all do the same thing and that's add functionality as well as a new way to interact with the platform. This is a fascinating idea as not many other platforms allow you to implement your own modifications. If you look at the Bots job postings on Discord's careers website, you'll see a big part of what they are trying to do with bots is “empower developers” and “give their dev community more ways to express their creativity and build exciting apps”. Let me be the first to tell you, this is exactly what they are doing. By giving people a way to interact with a piece of software that they use every day, it can inspire new and experience developers alike to share and develop their creativity. 43. Depop Bot - The Ultimate Guide (2020) Introduction 44. 20 Tips for Selling on Depop App: 2021 Edition Selling on Depop can be your full-time job and produce fairly good money if you take it seriously. Thank you for checking out the 44 most read stories about Bots on HackerNoon. to find the most read stories about any technology. Visit the /Learn Repo

44 Stories To Learn About Bots

In this modern world we live in, we no longer need any human assistance to create a voice that sounds exactly like another human would. Text-to-speech technology has dramatically transformed our reality, from assisting drivers navigating unfamiliar terrains with a GPS to helping visually impaired people in reading. Over the years, text-to-speech has made our lives significantly easier. Let’s stop for a moment and take a good look at how this technology came into play, and how it’s being revolutionized by AI right now. Early Days of Text-to-Speech Tech The first ventures into text-to-speech technology started in the mid-20th century when the first computer-based speech synthesis systems were created. These early systems were extremely rudimentary, with robotic voices that didn’t resemble real human speech very much – but they were comprehensible, which was a huge success in itself. Over the years, this technology was further developed and evolved, and today, we have easy access to tools that anyone can use to enhance their content. free text-to-speech The earliest text-to-speech systems used formant synthesis – a process that recreated human speech by synthesizing the basic components of sounds and putting them together in a harmonious order. Even though these systems sounded robotic and lacked many complexities of human speech, they were very efficient at creating aids for people who had trouble reading text. Nowadays, we don’t have to resort to these rudimentary techniques. In fact, text-to-speech technology has progressed so far that it’s now readily available to anyone with internet access, without requiring any technical skills. CapCut - a free online video and image editing software developed by the creators of TikTok – can create lifelike voices with ease, allowing users to select from a variety of templates and create voiceovers in many different languages, including English, Korean, Turkish, Spanish, Russian, German, Arabic, and more! The Challenges of Old TTS Technology and the Advent of AI One of the biggest challenges of the early text-to-speech systems was truthfully replicating human speech – the rich variation and intonation that goes into every spoken sentence. Our speech isn’t just a series of words. It has a rhythm, stress, pitch, and tone, which all carry emotional and meaningful information in addition to the words. Traditional TTS systems couldn’t replicate these complexities, resulting in flat, emotionless speech. Then, something new came along – Artificial Intelligence. With AI and deep learning models, artificial neural networks were designed to mimic human brain functioning. These networks helped create a new era of text-to-speech technology, where AI is used to learn and generate speech directly from text. AI-based text-to-speech takes advantage of massive amounts of data and sophisticated algorithms, generating incredibly realistic human speech with all its unique features. The algorithms train on existing databases of human speech, learning patterns, and subtleties similar to how a human would learn a language. First, the model is trained to understand phonetics and how different words are pronounced in various contexts. Then, the AI learns about capturing the right rhythm and intonation, implementing natural stress patterns that add emotions and meaning beyond bare words. Today, creating a realistic text-to-speech voiceover is as simple as writing the text and selecting a voice. CapCut, for example, provides a vast library of male and female voices to choose from, allowing users to select one that fits perfectly with their video. The speech rate and volume can be easily adjusted, creating accurate and realistic TTS in a matter of minutes. Text-to-speech isn’t the only AI-powered tool that CapCut offers. Users of the free online image and video editor can also take advantage of AI image style transfer, , AI image and video upscaling, photo colorizer, and AI-powered color correction. AI portrait generator With advances in artificial intelligence, editors no longer have to test and try different techniques – the AI will select the most suitable one on its own, enhancing images and videos effortlessly. The Future of Text-to-Speech Technology Today, text-to-speech technology no longer produces a dehumanized, lifeless voice that sounds like early 2000s synthesizers (remember Ivona?). With AI voices, even users without any technical knowledge can create a highly customizable voiceover, altering its speed, tone, accent, and many more aspects of the voice. These voices have tons of applications, from creating talking virtual assistants and accessibility aids to making audiobooks or video games without having to hire voice actors. As we move towards the future of TTS technology, we’ll be able to create more lifelike, expressive, and personalizable voices. Pretty soon, AI voiceovers may be indistinguishable from human speech, capable of conveying any emotion the author desires. This, of course, creates new issues that humanity will have to deal with – like the currently ongoing SAG-AFTRA (The Screen Actors Guild – American Federation of Television and Radio Artists) strikes that dispute studio usage of AI to recreate faces and voices of actors. This story was distributed as a release by Ascend under HackerNoon’s Brand As An Author Program. Learn more about the program here: https://business.hackernoon.com/brand-as-author

The Evolution of Text-to-Speech Tech: How Do AI Voices Work?

La evolución de la tecnología de texto a voz: ¿Cómo funcionan las voces de IA?

L'évolution de la technologie de synthèse vocale : comment fonctionnent les voix de l'IA ?

टेक्स्ट-टू-स्पीच तकनीक का विकास: एआई आवाज़ें कैसे काम करती हैं?

テキスト読み上げ技術の進化: AI 音声はどのように機能するのか?

A evolução da tecnologia de conversão de texto em fala: como funcionam as vozes de IA?

Sự phát triển của công nghệ chuyển văn bản thành giọng nói: Giọng nói AI hoạt động như thế nào?

文本转语音技术的演变：人工智能语音如何工作？

Do you know what generative voices are capable of? What vast possibilities does they open up, and how many areas can benefit from them? Maybe you need an app for voicing a video blog or book. Perhaps you want to bring a game character to life with a professional voiceover. How about creating an application for learning foreign languages? Today, has prepared a tutorial for you that will help you get to know AI voice technologies better! lablab.ai Let’s dive in! Introduction It  is one of the most exciting times for software development, what with the emergence of various " " tools in the market. Just name it, cover letter generation? Check! E-mail generation? Check! Automatic code comment generation? Check! Even outside coding and software development, the use case possibilities are enormous. generative AI Now, we can generate images with text prompts with various image generation models. Thus, it makes it possible for us to incorporate generated assets in our various products. The next question is: how about voices? The trend of user experiences in the past few years mentioned "voice command" as one of the emerging trend. It is only natural that the software we build will incorporate voices as one of the features. Which is why, in this tutorial, we will showcase the " " feature offered by in a simple app, which generates random words and have it spell it. To build the UI for this , we will use , a new UI library to share data science projects. Speech Synthesis ElevenLabs Python-based app Streamlit Introduction to ElevenLabs is a voice technology research company which offers solution. With easy to use , it allows developers to generate high-quality speeches using AI. It is made possible by the AI model which has been trained on a vast amount of audiobooks and also podcasts. The training allows the AI to deliver predictable and high-quality results in speech generation. ElevenLabs speech synthesis API There are two main features that has to offer, the first one is VoiceLab, where users can clone voices from recorded audio and/or existing pre-made voices, and also "design" voices based on gender, ages, ethnicities and races. Once users have the voices to work with, they can move on to the next feature, Speech Synthesis, where they can generate speeches using their designed voices or just using the pre-made ones. ElevenLabs Introduction to Model Anthropic's Claude is the latest AI model developed by , an AI research organization which focuses on improving the interoperability, robustness and safety of artificial intelligence systems. Claude Anthropic The model is designed to generate human-like responses, making it a powerful tool for a wide range of applications, from content creation, legal, to customer service. Just like any other in the market, is also trained on a diverse range of internet text. However, unlike most AI models, it has focus on "safety", which makes it possible to refuse outputs that it considers "harmful" or "untruthful" for the users. Claude AI models Claude Introduction to Streamlit is an open-source library that makes it easy and possible for developers and data scientists to create and share visually appealing and customized web apps. Developers can use to build and deploy fully featured data science apps in minutes. It is made possible by the simple and intuitive API that can be used to turn data scripts into UI components. Streamlit Python Streamlit Prerequisites Basic knowledge of and using Python UI development Streamlit Access to Anthropic API Access to ElevenLabs API Outline Initializing our Streamlit Project Adding Word Generation Feature using Model Claude Adding Speech Generation Feature using ElevenLabs API Testing the Word Generator App Discussion There are at least four steps that we will get through in this tutorial. First we need to initialize the Streamlit-based project, to get a general feel of developing user interfaces using Streamlit. Next, we start adding more features, beginning with engineering prompt to get model to give us a randomized word that is commonly misspelled. After that, we'll add text-to-voice generation provided by to demonstrate how the multilingual model spell the words. Finally, we're going to test the simple app. Claude ElevenLabs Initializing our Streamlit Project Let's get into the coding action! First, let's make a directory for our project and enter it! mkdir randomwords
cd randomwords Next, we're going to use this directory as the basis of our project. Because a project is essentially a project, we need to do some steps to initialize our project, such as defining and activating our virtual environment. Streamlit Streamlit Python Python # Creating the virtual environment
python -m venv env

# Activate the virtual environment
# On Linux/Mac
source env/bin/activate

# On Windows:
.\env\Scripts\activate Once activated, the output of our terminal should show the name of the virtual environment (env), like so: Next, it's time to install the libraries we need for this project! let's use the command to install the , , and library. Note that we also install a version-locked library to prevent a Pydantic-related error in one of the function. pip streamlit anthropic elevenlabs pydantic elevenlabs pip install streamlit anthropic elevenlabs "pydantic==1.*" With all the project's requirements out of the way, now let's dive into the coding part! Create a new file inside our project directory, let's call it . randomwords_app.py touch randomwords_app.py After the file is created, let's open the file in our favorite code editor or integrated development environment (IDE). For the starter, let's build our simple app from the simplest parts possible, a title and a text for the caption! import streamlit as st

st.title("Random Words Generator")

st.text("Hello, this is a random words generator app") To wrap up our project initialization part, let's try test running the app. Make sure that our current working directory is still inside our project and our virtual environment is already activated. When everything is ready, use the to run the app. streamlit run <app-name> streamlit run randomwords_app.py The app should open automatically in our default browsers! it should show the title and text for now. Next, we're going to add random word generation feature using model. Anthropic's Claude One last thing though, we'll have to provide our app with the API keys for the services that we're going to use, namely model and feature. As these keys are considered sensitive, we should keep them in a safe and isolated place. Anthropic's Claude ElevenLabs' Speech Synthesis However, this time we don't store them in a file. This is because Streamlit deal with environment variables differently. According to the , we need to create a secret configuration file inside a directory. We can create the directory inside our project and then create the file. .env documentation .streamlit mkdir .streamlit
touch .streamlit/secrets.toml Let's edit the TOML file we created, note that TOML file uses different formatting from the usual file. .env xi_api_key = "xxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
claude_key = "sk-ant-xxxxxxxxxxxxxxxxxxxxxxxxxxxxx" Adding Word Generation Feature using Claude Model In this step, we will add a button that will generate the random word, the heading element to show the generated word and the subheading to show the meaning of the word. However, coming from a webdev background, I strongly believe that UI elements should be placed and arranged inside containers. So, we'll do exactly that. Import the necessary libraries First of all, let's add some import statements. We're going to import the library to generate our random words. anthropic import streamlit as st
import anthropic Then, before we get to the UI part, let's create our word generation function first. Defining the word generation function def generate_word():
    prompt = (f"{anthropic.HUMAN_PROMPT} Give me one non-English word that's commonly misspelled and the meaning. Please strictly follow the format! example: Word: Schadenfreude; Meaning: joy at other's expenses."
              f"{anthropic.AI_PROMPT} Word: Karaoke; Meaning: a form of entertainment where people sing popular songs over pre-recorded backing tracks."
              f"{anthropic.HUMAN_PROMPT} Great! just like that. Remember, only respond following the pattern.")

    c = anthropic.Anthropic(api_key=st.secrets["claude_key"])
    resp = c.completions.create(
        prompt=f"{prompt} {anthropic.AI_PROMPT}",
        stop_sequences=[anthropic.HUMAN_PROMPT],
        model="claude-v1.3-100k",
        max_tokens_to_sample=900,
    )

    print(resp.completion)
    return resp.completion In this function, the most heavy lifting is done by ** **l (Thanks, ! 😉). However, our part in this function is how to make Claude return the exact format consistently. So we need to both instruct to "strictly follow the format" and give it an example response by adding it after our initial prompt. Anthropic's Claude mode Claude Claude Finally, we make sure that Claude comply with our agreements by ask it to "Remember to only respond following the pattern". The function ends by returning the response from . Claude Next, let's get back to editing the UI! Updating the UI st.title("Random Words Generator")

with st.container():
    st.header("Random Word")
    random_word = st.subheader("-")
    word_meaning = st.text("Meaning: -")

    st.write("Click the `Generate` button to generate new word")
    if st.button("Generate"):
        result = generate_word()
        # Split the string on the semicolon
        split_string = result.split(";")

        # Split the first part on ": " to get the word
        word = split_string[0].split(": ")[1]

        # Split the second part on ": " to get the meaning
        meaning = split_string[1].split(": ")[1]

        print(f"word result: {word}")
        random_word.subheader(word)
        word_meaning.text(f"Meaning: {meaning}") This time, we added a container with some elements inside it. The header, subheader for displaying the random word, and the text element to show the meaning of the word. We also have a text to show the hint on how to use the app, as well as a button beneath it. In , we can declare click event handler by using a conditional statement, where it returns when the button is clicked. In this code, we invoke the function which returns the generated word and the meaning, and split the result into separate variables for the word and the meaning, respectively. Finally, we update the subheader and the text element to display the word and the meaning. Streamlit True generate_word() Final form Let's double check our code once again! It should contains the import statements, the function for generating the random word, and the updated UI which contains subheader, and text elements as well as button that generate the word by invoking the function. generate_word() import streamlit as st
import anthropic

def generate_word():
    prompt = (f"{anthropic.HUMAN_PROMPT} Give me one non-English word that's commonly misspelled and the meaning. Please strictly follow the format! example: Word: Schadenfreude; Meaning: joy at other's expenses."
              f"{anthropic.AI_PROMPT} Word: Karaoke; Meaning: a form of entertainment where people sing popular songs over pre-recorded backing tracks."
              f"{anthropic.HUMAN_PROMPT} Great! just like that. Remember, only respond following the pattern.")

    c = anthropic.Anthropic(api_key=st.secrets["claude_key"])
    resp = c.completions.create(
        prompt=f"{prompt} {anthropic.AI_PROMPT}",
        stop_sequences=[anthropic.HUMAN_PROMPT],
        model="claude-v1.3-100k",
        max_tokens_to_sample=900,
    )

    print(resp.completion)
    return resp.completion


st.title("Random Words Generator")

with st.container():
    st.header("Random Word")
    random_word = st.subheader("-")
    word_meaning = st.text("Meaning: -")

    st.write("Click the `Generate` button to generate new word")
    if st.button("Generate"):
        result = generate_word()
        # Split the string on the semicolon
        split_string = result.split(";")

        # Split the first part on ": " to get the word
        word = split_string[0].split(": ")[1]

        # Split the second part on ": " to get the meaning
        meaning = split_string[1].split(": ")[1]

        print(f"word result: {word}")
        random_word.subheader(word)
        word_meaning.text(f"Meaning: {meaning}") Testing the Word Generation Function Let's run the app once again with the same command. We can also just rerun the app by clicking the upper right menu and click " " if we've had the app running before. Rerun It should show this updated user interface. Now, let's try clicking the button! Generate One of the sweet things about Streamlit is that it handled loading and provided the loading indicator out of the box. We should see the indicator in the upper-right corner, as well as the option to " " the operation. Neat, huh? stop After a few seconds, the result should be showed in the UI. Perfect! notice that the app correctly split the generated text from the model into word and the meaning. However, if the result doesn't come out according to the expected format, we can always click the button again. Claude Generate The next step is the main feature of this app, to incorporate speech generation into our random word generator. Besides demonstrating how to generate the audio file using the API provided by ElevenLabs, this step also serve to demonstrate the capabilities of ElevenLabs' multilingual model. Adding Speech Generation Feature using API ElevenLabs The first step of this section is, as you've probably guessed, is to add more import statement! So, let's add some functions from that we'll use for the speech generation feature. elevenlabs import streamlit as st
import anthropic
++ from elevenlabs import generate, set_api_key Next, we're going to define the function to handle the speech generation. def generate_speech(word):
    set_api_key(st.secrets['xi_api_key'])
    audio = generate(
        text=word,
        voice="Bella",
        model='eleven_multilingual_v1'
    )

    return audio Thanks to the simplicity and readability of , and also easy-to-use API, we can generate the speech by using this code alone! The function accepts the random word which we use to generate the speech. We also specifically use "eleven_multilingual_v1" model which is a multilingual model, perfect for our use case to demonstrate the spelling and pronounciation of foreign and commonly misspelled words! Finally, we use the " " voice for this tutorial, which is one of the pre-made voice provided by . Python ElevenLabs Bella ElevenLabs Next, we'll add an audio player to play the generated speech. print(f"word result: {word}")
    random_word.subheader(word)
    word_meaning.text(f"Meaning: {meaning}")
++     speech = generate_speech(word)
++     st.audio(speech, format='audio/mpeg') Just below our latest code from earlier, we add the variable to store the generated speech, and run the speech using audio player provided by function from Streamlit. At this point, our app should work as expected, only showing the audio player when there is a random word available to "read". st.audio Curious how it works out? me too! let's test the app now! Testing the Word Spelling Feature Let's run the app again using or just rerun the app if we have it running already. It should look exactly the same as the last time we left it. However, let's try to click the "Generate" button this time! streamlit run Amazing! this time, the app also shows an audio player! Let's try playing it. Using the multilingual model, the speech generated should use the accent and intonation which is close to the origin language of the word. For example, "entrepreneur" should be pronounced in French accent. Conclusion In this short tutorial, hopefully we've explored the capabilities of speech generation technology offered by . With the multilingual model, it's easy to generate speeches that is intended for non-English listener. In our use case, we need multilingual model to aid us in finding the correct way to pronounce and spell non-English words that are commonly misspelled. ElevenLabs With so many ideas, we invite developers to join us in creating the future! , in where you can create your own voice AI app with models! (Additionally, you can leverage other AI models such as large language models, image and video generative models, etc., as long as they are not in direct competition with the hackathon technology). On July 28 lablab.ai launching a challenge ElevenLabs *Your final submission should consist of a ready-to-play working prototype of your idea, a video pitch, and a presentation showcasing your solution. You can find more tutorials and you can other hackathons to build with cutting-edge technologies! HERE JOIN And big thanks to the - the Author of this article. 💚 Septian Adi Nugraha

Building a Simple Word Spelling App with ElevenLabs, Streamlit, and Claude

ElevenLabs, Streamlit, এবং Claude সহ একটি সহজ শব্দ বানান অ্যাপ তৈরি করা

Erstellen einer einfachen Wortbuchstabier-App mit ElevenLabs, Streamlit und Claude

Creación de una aplicación de ortografía de palabras simple con ElevenLabs, Streamlit y Claude

Créer une application d'orthographe simple avec ElevenLabs, Streamlit et Claude

इलेवनलैब्स, स्ट्रीमलिट और क्लाउड के साथ एक सरल शब्द वर्तनी ऐप बनाना

イレブンラボ、Streamlit、Claude を使用してシンプルな単語スペル アプリを構築する

ElevenLabs, Streamlit 및 Claude를 사용하여 간단한 단어 맞춤법 앱 구축

Construindo um aplicativo simples de ortografia de palavras com ElevenLabs, Streamlit e Claude

ElevenLabs, Streamlit ve Claude ile Basit Bir Kelime Yazımı Uygulaması Oluşturma

Xây dựng ứng dụng đánh vần từ đơn giản với ElevenLabs, Streamlit và Claude

使用 ElevenLabs、Streamlit 和 Claude 构建简单的单词拼写应用程序

is a lightweight and type-safe library to seamlessly create custom database repositories for applications. It provides developers with some built-in CRUD operations, enabling them to focus on adding their own operations, and ensures a clean separation between persistence and domain logic. Despite being database technology agnostic by nature, it currently includes an implementation for MongoDB. node-abstract-repository Node.js Note This article introduces , but it does not include all of its details. You may find the full documentation and some usage examples at , including an example of how to use this library in a NestJS-based application. node-abstract-repository this GitHub repository Why Build Yet Another Repository Library Joining made me switch my gears and start to develop JS/TS-based web applications. My very first assignment was connecting a small backend NestJS app with a database. We had big plans for our app and we soon realized that the domain model owned by my team was going to grow in complexity, and quickly. Furthermore, we saw that this was also the reality in other teams that wanted to use the same kind of solution. ClimatePartner MongoDB All considered, we decided to incorporate a database access utility that implements the to cleanly decouple persistence from domain logic while keeping query logic duplication to the bare minimum. We also wanted it to be of nature so that we could reuse the database access logic that is common to all custom repositories (i.e., CRUD operations over any persistent domain object). Finally, we needed it to implement the to support subtyping-based domain models. Repository pattern abstract Polymorphic pattern So, we started researching the state-of-the-art and discovered several libraries that could fit our requirements. The best candidates we were able to find were and . Mongoose is a well-known Node.js library for MongoDB that implements the Data Mapper pattern and lets developers define schemas to constrain the data models associated with their domain objects. Mongoose, Typegoose, TypeORM However, Mongoose works with concrete data models, which is a complex domain model scenario that results in query logic duplication. Typegoose is a type-safe Mongoose wrapper that allows schema constraint declaration at the domain object field level via JS decorators. Unfortunately, those very decorators leak persistence logic into the domain model. Besides, Typegoose also implements the Data Mapper pattern, thus sharing the same drawbacks of Mongoose. TypeORM, on another hand, implements the Repository pattern and provides some basic support for MongoDB. However, TypeORM presents several limitations compared to Mongoose. Although none of the existing alternatives were exactly fitting our needs, we saw value on Mongoose's solid, well-documented, and complete set of features. On the top of that, we wanted the most lightweight solution possible. That's why we decided to build . node-abstract-repository Cut to the Chase Say we have a domain model that specifies as the supertype domain object and and as subtypes. Here is one possible definition for that domain model: Book PaperBook AudioBook class Book implements Entity {
  readonly id?: string;
  readonly title: string;
  readonly description: string;
  readonly isbn: string;

  constructor(book: {
    id?: string;
    title: string;
    description: string;
    isbn: string;
  }) {
    this.id = book.id;
    this.title = book.title;
    this.description = book.description;
    this.isbn = book.isbn;
  }
}

class PaperBook extends Book {
  readonly edition: number;

  constructor(paperBook: {
    id?: string;
    title: string;
    description: string;
    isbn: string;
    edition: number;
  }) {
    super(paperBook);
    this.edition = paperBook.edition;
  }
}

class AudioBook extends Book {
  readonly hostingPlatforms: string[];
  readonly format?: string;

  constructor(audioBook: {
    id?: string;
    title: string;
    description: string;
    isbn: string;
    hostingPlatforms: string[];
  }) {
    super(audioBook);
    this.hostingPlatforms = audioBook.hostingPlatforms;
  }
} Now, you want to be able to persist and retrieve instances of and any of its subtypes from a MongoDB database. is a custom repository that specifies -related database operations. Here is its definition: Book MongooseBookRepository Book class MongooseBookRepository
  extends MongooseRepository<Book>
  implements BookRepository
{
  constructor() {
    super({
      Default: { type: Book, schema: BookSchema },
      PaperBook: { type: PaperBook, schema: PaperBookSchema },
      AudioBook: { type: AudioBook, schema: AudioBookSchema },
    });
  }

  async findByIsbn<T extends Book>(isbn: string): Promise<Optional<T>> {
    if (!isbn)
      throw new IllegalArgumentException('The given ISBN must be valid');
    return this.entityModel
      .findOne({ isbn: isbn })
      .exec()
      .then((book) => Optional.ofNullable(this.instantiateFrom(book) as T));
  }
} Voilà! inherits a series of CRUD database operations from and adds its own i.e., . Now you can simply instantiate and execute any of the database operations as follows: MongooseBookRepository MongooseRepository findByIsbn MongooseBookRepository const bookRepository = new MongooseBookRepository();
const books: Book[] = bookRepository.findAll(); No more leaking of persistence logic into your domain/application logic! Some Important Implementation Details You may be wondering about certain elements exposed in the previous example. Starting from the top part of the definition, the first of these is . This is one of the two main elements exposed at : a generic template that implements several common CRUD database operations. Furthermore, this class also handles the creation of a Mongoose data model capable of handling subtyping-based book data objects. All this complexity is hidden from you; all you need to know is that you have at your disposal to implement your own database operations, as shown at . MongooseBookRepository MongooseRepository<Book> node-abstract-repository entityModel findByIsbn The only extra effort that is required from you is that you define a map that specifies the types that compose your domain model and their related Mongoose schemas, as specified at the constructor of . It is also important to note that the key of the entry that specifies your domain object supertype must be named , and the key of each other entry must be named after the type name of the subtype domain object it refers to. This is a naming convention required by the inner implementation details of . If you do not have any domain object subtype in your domain model then simply instantiate a map with a unique entry that specifies a key and the type-schema values of your domain object. MongooseBookRepository Default MongooseRepository Default Another element that may catch your eye is . This is an optional interface that extends , the second main element exposed at : a database technology agnostic generic interface that specifies all the supported CRUD database operations to be implemented by any repository. The following snippet shows this relation between these two interfaces: BookRepository Repository node-abstract-repository interface BookRepository extends Repository<Book> {
  findByIsbn: <T extends Book>(isbn: string) => Promise<Optional<T>>;
} Handling instances of interface types instead of classes is a good practice in object-oriented applications; makes your code scalable and better suited to change. Therefore, I would highly recommend that you define your own repository interfaces, or in the case, you are only interested in exposing CRUD operations, instantiate your custom repositories as objects of type , where is your domain object supertype. depending on abstractions instead of implementations Repository<T> T On another hand, implements an interface called . This interface models any persistent domain object type. The interface defines an optional field. The optional nature of the field is due to the fact that its value is internally set by Mongoose. If you do not want your domain model to include this dependency, though, you can just ensure that your domain objects specify an field. And there is nothing wrong doing that, as is the canonical primary key in MongoDB documents. Book Entity id id?: string id Finally, many database operations return objects of type . This type is inspired in the type. Its purpose is to model the concept of "no result". This type is an excellent alternative to returning objects or throwing an exception. The usage of is safer and clearer than having to specify as result type or having to handle an exception that does not represent an application exceptional condition. So yes, this the feature that makes a bit opinionated, but I believe that is worth it. Optional Java Optional null Optional T | null node-abstract-repository Conclusion The library provides the means to  create custom repositories for Node.js-based applications in a fast and easy manner. It is conceived to achieve effective decoupling between persistence and domain logic. Its current implementation is a Mongoose wrapper that includes all the boilerplate code required to perform common CRUD operations over polymorphic domain objects in MongoDB. Thus, developers can focus in writing domain-specific database operations at custom repositories. node-abstract-repository Moreover, the library has been designed to allow developers use many interesting Mongoose features (e.g., creation and validation of polymorphic models as well as hook implementations, just two mention two) without having to deal with most of Mongoose's inner complexity. Also published . here Lead photo by on Unsplash Jan Antonin Kolar

Building a Scalable Node Abstract Repository for MongoDB

Stories, News & Mentions about AUTOBOOKS

Autobooks’s Stories on HackerNoon