Takoton 2
Tacotron 2 shine tsarin rubutu-zuwa-ƙarshe daga Google (2017) wanda ke juya rubutun da aka rubuta kai tsaye zuwa mel-spectrogram, wanda muryar murya ta jujjuya zuwa magana mai rai.
Dubawa
It produced audio rivaling human recordings on key benchmarks.
Zurfafa nutsewa
Tacotron 2 yana da manyan sassa biyu. Na farko, hanyar sadarwa ta jeri-zuwa-jeri tare da kulawa tana karanta haruffan rubutu kuma tana tsinkayar firam ɗin mel-spectrogram ta firam. Mai rikodin rikodin yana juya haruffa zuwa ɓoyayyun wakilci, tsarin kula da wuri mai dacewa yana daidaita rubutu zuwa firam ɗin sauti, kuma na'ura mai jujjuyawar atomatik yana fitar da sifofi yayin da alamar 'tsayawa' ke koya lokacin da furucin ya ƙare. Na biyu, ingantaccen vocoder na WaveNet yana canza wannan mel-spectrogram zuwa danyen igiyar igiyar ruwa. Ta hanyar rarrabuwar matsalar ta wannan hanyar, Tacotron 2 yana koyon prosody, lafazin magana, da taki daga bayanai tare da ƙaramin injiniyan hannu. Ya sami madaidaicin ma'aunin ra'ayi kusa da rikodin ƙwararru, yana mai da shi alamar ƙasa a cikin sautin sauti na halitta da samfuri don TTS na gaba.
Fahimtar Fasaha
mel-spectrogram shine hanyar sadarwa mai wayo tsakanin hanyoyin sadarwa guda biyu: yana da karamci kuma mai sauki ga samfurin kulawa don yin tsinkaya, duk da haka wadatar da mai sauti zai sake gina sauti mai inganci. Hankalin wuri-hankali yana hana gazawar gama gari kamar maimaitawa ko tsallake kalmomi ta yin la'akari da jeri-jefi na baya, da kuma na'ura mai jujjuyawar atomatik tare da alamar tasha koyo yana ba da samfurin damar sarrafa jimloli masu tsayi da kyau.
Dabarun Tasiri
Shiga ku isa
Yana inganta samun dama ta hanyar rubutu, ba da labari, da mu'amalar murya.
Kudin da kasafin kuɗi
Ƙungiyoyin kafofin watsa labaru na iya jigilar sauti mai gogewa cikin sauri tare da ƙaramin kasafin kuɗi.
Gudu da sikelin
Tsarin fuskantar abokin ciniki na iya aiwatar da hulɗar magana a mafi girman ma'auni.
Makomar Tacotton 2
Zane-zanen mataki biyu na Tacotron 2 ya yi wahayi zuwa ga guguwar TTS. Magada masu sauri waɗanda ba su da ƙarfi kamar FastSpeech 2 sun cire mai ƙididdigewa don sauri da kwanciyar hankali, kuma ana canza vocoder na WaveNet yanzu don HiFi-GAN ko samfuran watsawa. Filin yana motsawa zuwa cikakken ƙarshen-zuwa-ƙarshe da mai magana da yawa, bayyananniyar ra'ayi, da tsarin rufewar murya na sifili, amma Tacotron 2 ya kasance maƙasudin tushe na bututun da ke tushen spectrogram.
Aiwatar da Gaskiyar Duniya
Ƙarfafa sautin sauti na halitta a cikin samfuran rubutu-zuwa-magana da mataimaka na Google
Samar da bayyananniyar labari don littattafan mai jiwuwa da kwasfan fayiloli
Samar da muryoyi don masu karanta allo da software mai isa ga dama
Yin hidima azaman tushen bincike da misalin koyarwa don bututun TTS na jijiyoyi
Hatsari & Tsare-tsare
Rashin amfani da murya da haɗarin kwaikwaya yana ƙaruwa lokacin da aka rasa izini.
Daidaituwa na iya faɗuwa cikin lafuzza, yaruka, ko mahalli masu hayaniya.
Ana iya kuskuren sauti na roba don ingantacciyar magana ba tare da bayyananniyar lakabi ba.
Taswirar Hanya
Sami tabbataccen izini don ɗaukar murya, cloning, da sake amfani.
Gwajin ingantattun masu magana daban-daban da yanayin baya.
Ƙayyade lokacin da dole ne ɗan adam ya duba ko ya amince da abubuwan da aka fitar.
Yi lakabin sauti na roba da kuma adana bayanan da aka tabbatar don yin lissafi.
Ci gaba da Bincike
Free newsletter
Get the daily AI briefing
Three verified AI stories every weekday morning, written in plain English. Free forever, no ads.
One email each weekday. Unsubscribe in one click. We never sell or share your address.
Test yourself
Take the Tacotron 2 quiz
Instant feedback on every answer, and a shareable certificate with a verifiable ID once you pass a course.
Support free AI education. AI Understanding is a 501(c)(3) nonprofit — no ads, no paywall, ever. Make a donation
Jagora na gaba
Samfuran Samfura
Tambayoyin da ake yawan yi
What is Tacotron 2?
Tacotron 2 shine tsarin rubutu-zuwa-ƙarshe daga Google (2017) wanda ke juya rubutun da aka rubuta kai tsaye zuwa mel-spectrogram, wanda muryar murya ta jujjuya zuwa magana mai rai. Ya samar da rikodi na mutane masu kishiyantar sauti akan mahimmin ma'auni.
Wane matsakaicin wakilci Tacotron 2 ya annabta daga rubutu?
Tacotron 2's jerin-zuwa-jeri cibiyar sadarwa yana annabta mel-spectrogram, wanda vocoder sai ya juya zuwa sauti.
Menene ke canza Tacotron 2's spectrogram zuwa ainihin yanayin motsi?
Tacotron 2 yana haɗa ma'aunin hangen nesansa tare da ingantaccen vocoder WaveNet don samar da ingantaccen sauti na ƙarshe.
Wace matsala kulawa-matsala ke taimakawa hanawa?
Ta hanyar la'akari da jeri-jeri na baya, kulawar wuri-wuri yana rage gazawa kamar maimaita kalmomin da ake maimaitawa ko jefar da su.
Ta yaya Tacotron 2 zai san lokacin da ya daina haifar da furci?
Mai rikodin autoregressive yana tsinkayar alamar tsayawa, yana barin ƙirar ta sarrafa jimloli masu tsayi daban-daban.
Wane salon gine-gine gabaɗaya ke amfani da ɓangaren rubutu-zuwa-spectrogram?
Tacotron 2 yana amfani da ƙirar mai rikodin rikodin-zuwa jeri tare da kulawa ga haruffan taswira zuwa firam ɗin spectrogram.