Audio AI JAGORA

Source-Tace Vocoding da DUNIYA

Vocoder kayan aiki ne wanda ke raba magana cikin tubalan gininsa kuma ya sake gina shi.

2 min karatuAn sabunta ta ƙarshe

Dubawa

The source-filter model and the WORLD vocoder are classic methods that power text-to-speech and voice conversion by separating what your vocal cords do from what your mouth shapes.

Zurfafa nutsewa

Samfurin tace-tushen ya bayyana magana azaman guda biyu suna aiki tare: tushe (ƙarashin sautin muryar ku don sautin murya, ko iska mai hayaniya don raɗaɗi da baƙaƙe) ta hanyar tacewa (siffar makogwaro, baki, da hanci). Vocoder yana nazarin sautin da aka yi rikodi don kimanta waɗannan guda, sannan ya haɗa sabon sauti daga gare su. DUNIYA, wanda Masanori Morise ya sake shi a kusa da 2016, babban vocoder ne mai inganci wanda ke fitar da sigogi guda uku: F0 (madaidaicin madaurin tushe), ambulaf mai gani (tace, ta hanyar CheapTrick algorithm), da aperiodicity (nawa amo da sautin, ta hanyar PLATINUM/D4C). Ana iya canza waɗannan rafukan guda uku da kansu sannan a sake haɗa su, suna mai da DUNIYA dokin aiki don madaidaicin TTS da tsarin murya.

Fahimtar Fasaha

Ikon DUNIYA yana zuwa daga rabuwa mai tsabta. CheapTrick yana ƙididdige ambulaf mai santsi wanda ke da ƙarfi ga ƙananan kurakurai na F0, yayin da DIO/Harvest filin waƙa da D4C ke auna ƙarfin saƙo. Saboda farar sauti, katako, da hayaniya suna rayuwa a cikin rafukan ma'auni daban-daban, zaku iya matsawa F0 sama da octave ba tare da canza wanda sautin yake sauti ba, ko tsayin lokaci ba tare da canza sauti ba. Ƙwayoyin murya kamar WaveNet daga baya sun tsara tsarin motsi kai tsaye, amma DUNIYA ta kasance cikin sauri, fassarar, kuma ba ta da lasisi.

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 Tushen-Tace Vocoding da DUNIYA

Masu amfani da siginar sigina masu tsafta sun mamaye vocoders na jijiyoyi (HiFi-GAN, WaveRNN) don dabi'ar saman-ƙarshen, amma DUNIYA ba ta ɓace ba. Yana rayuwa a matsayin mai sauri, ƙarshen abokantaka na CPU a cikin bututun jujjuya murya, masu haɗar waƙa, da tushen bincike, kuma fasalulluka na F0-plus-spectral-envelope ɗin sa har yanzu suna ciyar da samfuran jijiya da yawa. Yi tsammanin tsarin haɗaɗɗiyar inda ma'auni masu fassara irin nau'in DUNIYA ke jagorantar dikodi na jijiyoyi, yana ba masu ƙirƙira daidaitaccen iko akan farar da katako ba tare da sadaukar da gaskiya ba.

Aiwatar da Gaskiyar Duniya

Kayan aikin musanya muryar da ke canza sautin lasifika da damara yayin da ake fahimtar magana

Rarraba masu haɗa murya (kamar yanayin yanayin UTAU/NNSVS) waɗanda ke sake haɗa bayanin kula a sabbin filaye.

Tsarukan rubutu-zuwa-magana waɗanda ke haifar da F0, spectral, da rafukan aperiodicity kafin yin magana.

Tushen bincike na magana don jujjuyawar sauti, shimfiɗa lokaci, da gyaran gyare-gyare ba tare da sake horarwa ba

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

1

Sami tabbataccen izini don ɗaukar murya, cloning, da sake amfani.

2

Gwajin ingantattun masu magana daban-daban da yanayin baya.

3

Ƙayyade lokacin da dole ne ɗan adam ya duba ko ya amince da abubuwan da aka fitar.

4

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 Source-Filter Vocoding and WORLD quiz

Instant feedback on every answer, and a shareable certificate with a verifiable ID once you pass a course.

Fara tambayoyi

Support free AI education. AI Understanding is a 501(c)(3) nonprofit — no ads, no paywall, ever. Make a donation

Jagora na gaba

Demucs Music Source Rabe

Tambayoyin da ake yawan yi

What is Source-Filter Vocoding and WORLD?

Vocoder kayan aiki ne wanda ke raba magana cikin tubalan gininsa kuma ya sake gina shi. Samfurin tace-tushen da kuma muryar DUNIYA hanyoyi ne na yau da kullun waɗanda ke ba da ƙarfin rubutu-zuwa-magana da canza murya ta hanyar raba abin da igiyoyin muryar ku ke yi da abin da bakinku ya siffata.

A cikin samfurin-tace tushen magana, menene 'tace' ke wakilta?

Tace ita ce sautin muryar murya - sifar makogwaro, baki, da hanci waɗanda ke canza sautin. Madogararsa ita ce kurwar murya ko ƙarar iska.

Wadanne ma'auni guda uku ne muryar DUNIYA ta fitar daga magana?

DUNIYA tana lalata magana zuwa mitar asali (F0), ambulaf mai ban mamaki (timbre/tace), da aperiodicity (amo da ma'aunin sauti).

Menene sunan Algorithm na DUNIYA don ƙididdige ambulan na gani?

CheapTrick yana ƙididdige ambulaf mai santsi mai santsi wanda ke da ƙarfi ga ƙananan kurakurai a cikin ƙimar farar.

Me yasa raba farar daga ambulan na gani yana da amfani?

Saboda farar (F0) da timbre (ambulan bakan) rafukan rafuka ne masu zaman kansu, zaku iya ɗaga ko rage farar yayin adana asalin lasifika.

Ta yaya DUNIYA ke kwatanta da na'urar vocoders kamar HiFi-GAN?

DUNIYA siginar murya ce mai sarrafa sigina: mai sauri, mai fassara, da abokantaka na CPU. Vocoders na jijiyoyi yawanci suna samun mafi girman halitta amma sun fi nauyi.