Ndzi na khume ra malembe ndzi ri na iPhone, naswona ndza yi tsakela. Ku hambana ni vanhu van’wana, ndzi yi tsakela swinene Siri naswona ndzi yi tirhisa nkarhi na nkarhi. Kambe endzhaku ka khume ra malembe, Siri a nga si swi vona leswaku loko yi tsala matsalwa ya mina, yi fanele yi swi tiva leswaku vito ra nsati wa mina a hi Aaron, i Erin. Ndzi rivalela ku tirhisiwa ka mbulavulo ku ya eka tsalwa, loku lavaka switirhisiwa swo tala, kambe endzhaku ka loko ndzi lulamise xihoxo xexo kan’we ivi ndzi rhumela tsalwa leri pfuxetiweke, ku lulamisiwa koloko a ku fanele ku hlayisiwile eka matimu ya ku lulamisa eka riqingho ra mina—fayili leyitsongo leyi tirhisiwaka hi poso -processing transformer model, kun’we na swikombiso swin’wana, ku endla leswaku xihoxo lexi xi nga ha endleki ngopfu. Ndza swi tiva leswaku ku vitana ntirho wa mbulavulo wa iPhone eka matsalwa Siri swi olovisa ngopfu, kambe hi yona ndlela leyi vana va mina va ehleketaka ha yona hi ‘AI eka iPhone ya mina.’  Maendlelo ya mbulavulo ku ya eka tsalwa hakanyingi ma lwisana ni ti- homophone—marito lama twalaka ma fana kambe ma ri ni mapeletelo ni tinhlamuselo leti hambaneke. Swihoxo leswi swi nga ha ku heta matimba, ngopfu-ngopfu loko swi khumba mavito ya munhu kumbe marito lama tirhisiwaka ngopfu. Xilotlelo xo lulamisa xiphiqo lexi a hi ku pfuxeta njhini yo lemuka mbulavulo kambe eka leyara yo vevuka, yo lulamisa matsalwa ya le ndzhaku ka ku tsariwa leyi pfumelelanaka na ku lulamisiwa ka mutirhisi hi ku famba ka nkarhi. Hi leyi khodi leyi sekeriweke eka PyTorch leyi ndzi yi endleke ku lulamisa leswi.  Yi super compact naswona ya olova ku yi tirhisa eka foni endzhaku ko hlengeleta eka mobile. Ndza swi tiva leswaku endzhaku ka Siri ku na sete leyi rharhanganeke swinene ya timodeli leti bohiweke hi tinketana, kutani khodi leyi yi nga tirhisiwa ntsena ku nyika xivumbeko lexintshwa tanihi ku nghenisa eka timodeli teto, xikoro lexi pfunaka ku endla leswaku ku tsariwa ka munhu hi xiyexe loko ku humelela tihomophone to karhi. Kambe swinga olova ku tirhisa leswi tani hi post processing layer.  Leswi a swi bohi ku rindza ku humesiwa ka riqingho lerintshwa ku tirhisiwa. Swi ta endla leswaku vutomi byi antswa eka mina eka update leyi landzelaka leyi Apple yi yi humesaka eka iPhone ya mina.   Miehleketo ya Nkoka  Endlelo leri ri kongomisa eka swiyenge swinharhu leswikulu:    Yi hlayisa ku lulamisiwa ka mutirhisi wa khale, yi rhangisa marito lawa mutirhisi a ma lulamiseke hi ku kongoma khale. Matimu ya Ndzulamiso:    Ku landzelela marito kumbe mavito lama tirhisiwaka ngopfu, ku avela ku koteka lokukulu eka lawa ya tirhisiwaka ngopfu. Ku Tihlanganisa Nkarhi Na Nkarhi:    Yi tirhisa Vukorhokeri bya Ririmi ra Ntumbuluko (NLP) ku xopaxopa tsalwa leri rhendzeleke ku kuma swikombiso leswi pfunaka ku hambanyisa tihomophone. Nxopaxopo wa Mongo:  Sisiteme yi hlayela xibalo xa ku koteka eka muhlawuriwa un’wana na un’wana wa homophone hi ku ya hi swilo leswi swinharhu naswona yi hlawula ku lulamisiwa loku nga kotekaka swinene. Laha hansi ku na ku tirhisiwa ka Python loku hambanisiweke hi swiyenge leswi nga na tinhlamuselo.   Ku layicha Dathabeyisi ya Tihomofoni  Goza ro sungula i ku tumbuluxa kumbe ku layicha database ya tihomophone. Lawa i mimpatswa ya marito (kumbe mintlawa) leyi nga ha pfilunganyeka hi nkarhi wa ku tsariwa.   # Homophones database homophones_db = { "Aaron": ["Erin"], "bare": ["bear"], "phase": ["faze"], "affect": ["effect"], }  Lexi i xihlamusela-marito xo olova laha xilotlelo ku nga rito leri tsariweke hi ndlela leyi hoxeke, naswona nkoka i nxaxamelo wa swin’wana swa homophone. Xikombiso, "phase" yi nga pfilunganyeka na "faze". Endzhaku, database leyi yi ta vutisisiwa loko ku hlangana na rito leri nga twisisekiki.   Ku Landzelela Matimu ya Ndzulamiso  Khodi yi landzelela ku lulamisiwa ka mutirhisi eka xihlamusela-marito laha xilotlelo xin’wana na xin’wana xi nga tuple ya (original_word, corrected_word) naswona nkoka i nhlayo ya minkarhi leyi mutirhisi a lulamiseke xihoxo xexo.  Mulandzeri wa matimu ya ku lulamisiwa   # Correction history tracker correction_history = { ("phase", "Faye's"): 3, ("bear", "bare"): 2, }  Loko mutirhisi a lulamisa "phase" eka "Faye's" kanharhu, sisiteme yi rhangisa ku lulamisiwa loku eka matsalwa ya nkarhi lowu taka.   Ku Tihlanganisa Nkarhi Na Nkarhi  Xin’wana lexi kucetelaka ku hlawuriwa ka homophone i leswaku rito ro karhi ri tirhisiwa kangani. Leswi swi nga ha va mavito ya munhu hi xiyexe kumbe marito lawa mutirhisi a talaka ku ma thayipa.   # Frequent contact tracker frequent_contacts = { "faye": 15, "phase": 5, "erin": 10, "aaron": 2, }  Endlelo leri ri nyika ntikelo lowukulu eka marito lama tirhisiwaka ngopfu loko ku hambanisiwa tihomophone. Xikombiso, loko "faye" yi humelela minkarhi ya 15 kambe "phase" yi humelela minkarhi ya 5 ntsena, "faye" yi ta tsakeriwa.   Nxopaxopo wa Mongo  Swikombiso swa mongo swi humesiwa eka xivulwa lexi rhendzeleke ku ya emahlweni ku antswisiwa nhlawulo. Xikombiso, loko xivulwa xi ri na risivi "yena", sisiteme yi nga ha tsakela "Erin" ku tlula "Aaron". ku suka eka titransformer ku nghenisa phayiphi  Layicha modele wa NLP wa nxopaxopo wa xiyimo   from transformers import pipeline # Load an NLP model for context analysis context_analyzer = pipeline("fill-mask", model="bert-base-uncased") def detect_context(sentence): """Detect context-specific clues in the sentence.""" pronouns = ["he", "she", "his", "her", "their"] tokens = sentence.lower().split() return [word for word in tokens if word in pronouns]  Ntirho lowu wu skena xivulwa ku kuma maviti lama kongomisiweke eka rimbewu kumbe swikombiso swin’wana leswi nga kombisaka nhlamuselo leyi kunguhatiweke ya rito.  Ku hlayela Swikoweto swa ku koteka  Mukamberiwa un’wana na un’wana wa homophone u nyikiwa xibalo xa ku koteka lexi simekiweke eka:    : Ndzilo wa le henhla (xikombiso, 3x). Switsundzuxo swa Nkarhi lowu hundzeke    : Ndzilo wa le xikarhi (xikombiso, 2x). Ku tirhisiwa ka nkarhi na nkarhi    : Ku hunguta ntiko (xikombiso, 1x). Ku fambelanisa mongo   def calculate_likelihood(word, candidate, sentence): """Calculate a likelihood score for a homophone candidate.""" correction_score = correction_history.get((word, candidate), 0) * 3 frequency_score = frequent_contacts.get(candidate, 0) * 2 context = detect_context(sentence) context_clues = homophones_db.get(candidate, []) context_score = sum(1 for clue in context if clue in context_clues) return correction_score + frequency_score + context_score  Xikoro lexi xi hlanganisa swilo leswinharhu ku kumisisa homophone leyi nga ha vaka kona.  Ku Hluvukisa Tihomophone  Hi swikoweto swa ku koteka leswi hlayiweke, sisiteme yi hlawula homophone leyi nga na swikoweto swa le henhla.   def prioritize_homophones(word, candidates, sentence): """Prioritize homophones based on their likelihood scores.""" likelihoods = { candidate: calculate_likelihood(word, candidate, sentence) for candidate in candidates } return max(likelihoods, key=likelihoods.get) def disambiguate_homophone(word, sentence): """Disambiguate homophones using likelihood scores.""" candidates = homophones_db.get(word, []) if not candidates: return word return prioritize_homophones(word, candidates, sentence)  Endlelo leri ri tiyisisa leswaku rito leri faneleke swinene ri hlawuriwa hi ku ya hi matimu, nkarhi na nkarhi na mongo.   Ku Endliwa ka Matsalwa lama heleleke  Endlelo ri tirhisa xivulwa hinkwaxo, ri tirhisa logic ya disambiguation eka rito rin’wana na rin’wana.   def process_transcription(transcription): """Process the transcription to correct homophones.""" words = transcription.split() corrected_words = [disambiguate_homophone(word, transcription) for word in words] return " ".join(corrected_words)  Xikombiso lexi heleleke xa Mafambiselo ya Ntirho   # Example transcription and correction raw_transcription = "This is phase one plan." corrected_transcription = process_transcription(raw_transcription) print("Original Transcription:", raw_transcription) print("Corrected Transcription:", corrected_transcription) # Simulate user feedback update_correction_history("phase", "faye") print("Updated Correction History:", correction_history) print("Updated Frequent Contacts:", frequent_contacts)  Ku pfuxeta Vuyelo  Loko mutirhisi a lulamisa xihoxo, matimu ya ku lulamisa na ku tihlanganisa nkarhi na nkarhi swa pfuxetiwa ku antswisa ku vhumbha ka nkarhi lowu taka.   def update_correction_history(original, corrected): """Update correction history and frequent contacts.""" correction_history[(original, corrected)] = correction_history.get((original, corrected), 0) + 1 frequent_contacts[corrected] = frequent_contacts.get(corrected, 0) + 1 frequent_contacts[original] = max(0, frequent_contacts.get(original, 0) - 1)  Xikombiso xa ku tsariwa ka matsalwa na ku lulamisa   Original Transcription: This is phase one plan. Corrected Transcription: This is Faye's one plan. Updated Correction History: {('phase', 'Faye's'): 4} Updated Frequent Contacts: {'Faye's': 16, 'phase': 4}   Mahetelelo  Leyara leyi yo vevuka ya ku lulamisa matsalwa yi ndlandlamuxa ku pakanisa ka matirhiselo ya mbulavulo ku ya eka tsalwa hi ku dyondza eka ku lulamisiwa ka vatirhisi, ku tirhisa matirhiselo ya nkarhi na nkarhi, na ku xopaxopa mongo. Yi ringanile ku ringana ku tirha eka switirhisiwa swa tiselfoni naswona yi cincacinca eka swilaveko swa mutirhisi un’wana na un’wana, yi nyika ndlela yin’wana yo tlhariha eka timodeli ta ndhavuko ta static. Hi matshalatshala lamantsongo, Apple—kumbe khamphani yin’wana na yin’wana—yi nga hlanganisa ntirho lowu ku endla leswaku vapfuni va xiviri vo fana na Siri va hlamula swinene na ku va va munhu hi xiyexe.

The code in this story is for educational purposes. The readers are solely responsible for whatever they build with it.

Read My Stories

Audio leyi yi humesiwile hi ririmi ro sungula ra xitori!

U karhele Hi Siri Ku Pfilunganya Vito Ra Wena? Hi leyi Ndzulamiso wo Olova

About Author

SWIVUTISO SWA SWIVUTISO

HANG TAGS

XITLHOKOVETSELO LEXI XI NYIKERIWE EKA

Related Stories

Digg Free Gear? – A Thank You Contest by HackerNoon

Digg Free Gear? – A Thank You Contest by HackerNoon

Light-Mode

Classic

Newspaper

Dark-Mode

Neon Noir

Minty

HN StartUps