UMHLAHLANDLELA WE-AI womsindo

I-FastPitch Pitch-Controllable TTS

I-FastPitch imodeli esheshayo, engaguquki yokuguqula umbhalo ube inkulumo ebikezela ngokusobala ukuphakama (imvamisa eyisisekelo) yawo wonke amathokheni okufakwayo, ekuvumela ukuthi uhlele iphimbo nokugcizelela ngokumane ukale lezo zibikezelo.

Uhlolojikelele

I-FastPitch imodeli esheshayo, engaguquki yokuguqula umbhalo ube inkulumo ebikezela ngokusobala ukuphakama (imvamisa eyisisekelo) yawo wonke amathokheni okufakwayo, ekuvumela ukuthi uhlele iphimbo nokugcizelela ngokumane ukale lezo zibikezelo. Ibalulekile ngoba ikhiqiza i-mel-spectrogram egcwele ngokuhambisana - eshesha kakhulu kunamamodeli amadala alandelanayo - kuyilapho inikeza ukulawula okuqondile, okuhumusekayo kumculo wezwi.

I-FastPitch Pitch-Controllable TTS ihlezi ku-audio-AI workflows eguqula inkulumo, umculo, nomsindo wokuxhumana, ukufinyeleleka, nokukhiqizwa kwemidiya.

I-Deep Dive

I-FastPitch, eyethulwe yi-NVIDIA ngo-2020, yakhela ekwakhiweni okufanayo kwe-FastSpeech ngokwengeza isibikezelo sephimbo esicacile. Kufonimu ngayinye yokokufaka noma uhlamvu ibikezela inani elilodwa lemvamisa eyisisekelo, bese ibeka idekhoda ye-mel-spectrogram kuleyo khokhoni yephimbo. Ngenxa yokuthi iphimbo liyisignali ehlukile, efundeka umuntu, ungakwazi ukuliphindaphinda, ulisuse, noma ulihlele ngesandla ngaphambi kokuhlanganiswa ukuze ushintshe ukugcizelela, wenze inkulumo izwakale bukhoma, noma ulungise ukulethwa okuyisicaba - ngaphandle kokuqeqeshwa kabusha. I-spectrogram yonke ikhiqizwa ngokudlula okukodwa okuya phambili (okungeyona i-autoregressive), ngakho-ke isizukulwane sicishe sihlele ubukhulu ngokushesha kunamamodeli we-autoregressive afana ne-Tacotron 2, futhi iphimbo elibikezelwe libuye lithuthukise imvelo iyonke.

I-Technical Insight

I-FastPitch ilinganisa imvamisa yeqiniso eliphansi eliyisisekelo phakathi nobude bethokheni ngayinye phakathi nokuqeqeshwa, ngakho isibikezeli sifunda inani lephimbo elilodwa ngophawu ngalunye kunohlaka ngalunye - okwenza isilawuli sibe mahholo kodwa sibe esinembile. Uma kucatshangelwa, lokho kuphakama kwethokheni ngayinye kusakazwa phakathi nesikhathi esibikezelwe sethokheni futhi kwengezwe njengesignali yokubeka isimo kusikhiphi esisekelwe ku-transformer. Ngenxa yokuthi ayikho iluphu yempendulo ye-autoregressive, wonke amafreyimu okukhiphayo abalwa ngesikhathi esisodwa ku-hardware efanayo, asusa ukunqwabelana kwamaphutha kanye nesivinini esinensayo samadekhoda esinyathelo nesinyathelo.

Ukwenza kahle i-FastPitch Pitch-Controllable TTS

I-FastPitch imodeli esheshayo, engaguquki yokuguqula umbhalo ube inkulumo ebikezela ngokusobala ukuphakama (imvamisa eyisisekelo) yawo wonke amathokheni okufakwayo, ekuvumela ukuthi uhlele iphimbo nokugcizelela ngokumane ukale lezo zibikezelo. Ibalulekile ngoba ikhiqiza i-mel-spectrogram egcwele ngokuhambisana - eshesha kakhulu kunamamodeli amadala alandelanayo - kuyilapho inikeza ukulawula okuqondile, okuhumusekayo kumculo wezwi. I-FastPitch Pitch-Controllable TTS ihlezi ku-audio-AI workflows eguqula inkulumo, umculo, nomsindo wokuxhumana, ukufinyeleleka, nokukhiqizwa kwemidiya. Ukuze wakhe ukuqonda okujulile, phatha i-FastPitch Pitch-Controllable TTS njengemodeli yokusebenza, hhayi isici esisodwa: chaza imiphumela efiselekayo, ucacise ukucabanga, futhi uhlukanise lokho uhlelo olungakwenza ngokwethembeka kulokho okusadinga ukwahlulela kochwepheshe.

Empeleni, amaqembu aqinile asebenzisa i-FastPitch Pitch-Controlable TTS aphatha ikhwalithi, ukubambezeleka, kanye nemvume njengezingxenye ezibaluleke ngokulinganayo zesu lokuthunyelwa. Babhala imibandela yempumelelo ecacile, ukuhlola okuqhathaniswa nedatha engokoqobo nokugeleza komsebenzi, futhi baphindaphinde ngokusekelwe kumaphethini okuhluleka aqashiwe esikhundleni sokuwina kwebhentshimakhi yesikhathi esisodwa. Yilapho ukuqonda kwethiyori kuguquka kube amandla ahlala njalo kuwo wonke umkhiqizo, inqubomgomo, kanye nokusebenza.

Ithuthukisa ukufinyeleleka ngokuloba, ukulandisa, nezixhumi ezibonakalayo zezwi. Ngesikhathi esifanayo, ukusetshenziswa kabi kwezwi kanye nezingozi zokuzenza ongeyena ziyakhuphuka uma imvume ingekho. Indlela eqine kakhulu iwukuhlanganisa isivinini sokuhlola nesiyalo sokuphatha: qhuba abashayeli bezindiza, bamba ubufakazi, ushicilele amalogi ezinqumo, futhi ubuyekeze izivikelo ngokuqhubekayo njengoba imodeli yokuziphatha, okulindelwe ngabasebenzisi, kanye nezimfuneko zokulawula zishintsha.

I-Strategic Impact

Ithuthukisa ukufinyeleleka ngokuloba, ukulandisa, nezixhumi ezibonakalayo zezwi.

Ithuthukisa ukufinyeleleka ngokuloba, ukulandisa, nezixhumi ezibonakalayo zezwi. Ekusetshenzisweni kwekhwalithi ephezulu, lokhu kuhunyushwa emithethweni yokusebenza elinganisekayo, imingcele yobunikazi, nemikhuba yokubuyekeza ephindelelayo ukuze amaqembu akwazi ukukala ukuzethemba esikhundleni sokukala ukungaqondakali.

Amaqembu emidiya angathumela umsindo opholishiwe ngokushesha ngamabhajethi amancane.

Amaqembu emidiya angathumela umsindo opholishiwe ngokushesha ngamabhajethi amancane. Ekusetshenzisweni kwekhwalithi ephezulu, lokhu kuhunyushwa emithethweni yokusebenza elinganisekayo, imingcele yobunikazi, nemikhuba yokubuyekeza ephindelelayo ukuze amaqembu akwazi ukukala ukuzethemba esikhundleni sokukala ukungaqondakali.

Amasistimu abhekene nekhasimende angacubungula ukusebenzelana okukhulunyiwe ngesilinganiso esikhulu.

Amasistimu abhekene nekhasimende angacubungula ukusebenzelana okukhulunyiwe ngesilinganiso esikhulu. Ekusetshenzisweni kwekhwalithi ephezulu, lokhu kuhunyushwa emithethweni yokusebenza elinganisekayo, imingcele yobunikazi, nemikhuba yokubuyekeza ephindelelayo ukuze amaqembu akwazi ukukala ukuzethemba esikhundleni sokukala ukungaqondakali.

Ikusasa le-FastPitch Pitch-Controllable TTS

Ifilosofi yokulawula okubekela sobala ye-FastPitch ithonya amasistimu amasha aveza amandla, ubude besikhathi, nomuzwa njengamasiginali ahlelekayo eduze kwephimbo, okunikeza abadali isixhumi esibonakalayo sebhodi lokuxutshwa lezwi. Lindela ukuhlanganiswa okuqinile namavokhoda e-neural afana ne-HiFi-GAN kumaphayiphi esikhathi sangempela asuka ekupheleni ukuya ekupheleni, ukulawulwa kwephimbo kohlaka olungcono kakhulu lokuhlanganiswa kokucula, nokuhlukahluka kwezilimi eziningi nezikhulumi eziningi. Njengoba i-TTS elawulekayo isabalala ezinhlelweni zokusebenza ezibukhoma, ukubambezeleka okuphansi kudivayisi ukusetshenziswa kanye nokudluliselwa kwesitayela okucacile kuzoba izikhombisi-ndlela ezinkulu.

Ukuqaliswa Komhlaba Wangempela

Ukuvumela abaklami bomsizi wezwi bakhulise izwi kumagama angukhiye ukuze izimpendulo ezikhulunywayo zizwakale zigcizelela kakhulu

Ukukhiqiza ukucula noma inkulumo emnandi ngokuhlela ngesandla ubuningi obuyisisekelo benothi ngalinye

Ukulandisa kwesikhathi sangempela kumathuluzi adinga imigqa eminingi ehlanganiswe ngokushesha ngenxa yokuqoshwa kwayo okufanayo

Ukulungisa ukulethwa okuyisicaba noma okurobhothi kuzimemezelo ezihlanganisiwe ngokukala i-pitch contour ebikezelwe

Amaphethini Okusebenzisa

I-FastPitch Pitch-Controllable TTS isebenza

Ukuvumela abaklami bomsizi wezwi bakhulise izwi kumagama angukhiye ukuze izimpendulo ezikhulunywayo zizwakale zigcizelela kakhulu.

Ukuvumela abaklami bomsizi wezwi bakhulise izwi kumagama ayisihluthulelo ukuze izimpendulo ezikhulunywayo zizwakale zigcizelela kakhulu Amathimba ngokuvamile athola imiphumela engcono lapho echaza ikhwalithi ephezulu ngaphambili, egcina indlela yokukhuphuka yomuntu yamakesi asemaphethelweni, futhi alandelele kokubili izinzuzo zokukhiqiza nezindleko zamaphutha ngokuhamba kwesikhathi.

I-FastPitch Pitch-Controllable TTS isebenza

Ukukhiqiza ukucula noma inkulumo emnandi ngokuhlela ngesandla ubuningi obuyisisekelo benothi ngalinye.

Ukukhiqiza ukucula noma inkulumo emnandi ngokuhlela ngesandla imvamisa yenothi ngalinye Amaqembu ngokuvamile athola imiphumela engcono uma echaza izinga eliphezulu ngaphambili, egcina indlela yokukhuphuka yabantu yamakesi asemaphethelweni, futhi elandelela kokubili izinzuzo zokukhiqiza nezindleko zamaphutha ngokuhamba kwesikhathi.

I-FastPitch Pitch-Controllable TTS isebenza

Ukulandisa kwesikhathi sangempela kumathuluzi adinga imigqa eminingi ehlanganiswe ngokushesha ngenxa yokuqoshwa kwayo okufanayo.

Ukulandisa kwesikhathi sangempela kumathuluzi adinga imigqa eminingi edidiyelwe ngokushesha ngenxa yokuqoshwa kwayo okufanayo Amathimba ngokuvamile athola imiphumela engcono lapho echaza izinga eliphezulu ngaphambili, egcina indlela yokukhuphuka yomuntu yamakesi asemaphethelweni, futhi elandelela kokubili izinzuzo zokukhiqiza nezindleko zamaphutha ngokuhamba kwesikhathi.

I-FastPitch Pitch-Controllable TTS isebenza

Ukulungisa ukulethwa okuyisicaba noma okurobhothi kuzimemezelo ezihlanganisiwe ngokukala i-pitch contour ebikezelwe.

Ukulungisa ukulethwa okuyisicaba noma okunerobhothi ezimemezelweni ezihlanganisiwe ngokukala i-pitch contour ebikezelwe Amaqembu ngokuvamile athola imiphumela engcono uma echaza ikhwalithi ephezulu ngaphambili, egcina indlela yokukhuphuka yabantu yamakesi asemaphethelweni, futhi elandelela kokubili izinzuzo zokukhiqiza nezindleko zamaphutha ngokuhamba kwesikhathi.

Izingozi & Guardrails

!

Ukusetshenziswa kabi kwezwi kanye nezingozi zokuzenza ongeyena ziyanda uma imvume ingekho.

!

Ukunemba kungase kwehle kuzo zonke izinhlobo zokuphimisela, izilimi zesigodi, noma izindawo ezinomsindo.

!

Umsindo wokwenziwa ungenziwa iphutha njengenkulumo eyiqiniso ngaphandle kokulebula okucacile.

Ukuqalisa Umhlahlandlela

1

Thola imvume esobala yokuthwebula izwi, ukuhlanganisa, nokusebenzisa kabusha.

Thola imvume esobala yokuthwebula izwi, ukuhlanganisa, nokusebenzisa kabusha. Phatha isinyathelo ngasinye njengesango lobufakazi: uma imibandela ingafinyelelwa, misa ukukhishwa, vala igebe, bese unweba ukusetshenziswa.

2

Ikhwalithi yokuhlola kuzo zonke izipikha nezimo zangemuva.

Ikhwalithi yokuhlola kuzo zonke izipikha nezimo zangemuva. Phatha isinyathelo ngasinye njengesango lobufakazi: uma imibandela ingafinyelelwa, misa ukukhishwa, vala igebe, bese unweba ukusetshenziswa.

3

Chaza ukuthi kunini lapho umuntu kufanele abuyekeze noma agunyaze okuphumayo.

Chaza ukuthi kunini lapho umuntu kufanele abuyekeze noma agunyaze okuphumayo. Phatha isinyathelo ngasinye njengesango lobufakazi: uma imibandela ingafinyelelwa, misa ukukhishwa, vala igebe, bese unweba ukusetshenziswa.

4

Lebula umsindo wokwenziwa futhi ugcine amarekhodi atholakalayo ukuze aziphendulele.

Lebula umsindo wokwenziwa futhi ugcine amarekhodi atholakalayo ukuze aziphendulele. Phatha isinyathelo ngasinye njengesango lobufakazi: uma imibandela ingafinyelelwa, misa ukukhishwa, vala igebe, bese unweba ukusetshenziswa.

Qhubeka Uhlole