I-AI Foundations
Qonda ukuthi iyini i-AI, ukuthi amasistimu afunda kanjani, lapho ehluleka khona, kanye nendlela yokwahlulela izimangalo ngaphandle kokukhohlisa.
Umtapo wezincwadi we-AI wamahhala
117 imihlahlandlela yesiNgisi esicacile, izindlela zokufunda ezihlelekile, nomtapo wolwazi ovulekile — wakhiwe inhlangano engenzi nzuzo ye-501(c)(3) ezimele ukuze noma ubani aqonde i-AI yesimanje.
Qala lapha
Isifundo ngasinye sihlanganisa imiphumela esobala, amakhono afakwe kumephu, imisebenzi yokuzilolonga, kanye netshe eliyinhloko elisetshenzisiwe.
Qonda ukuthi iyini i-AI, ukuthi amasistimu afunda kanjani, lapho ehluleka khona, kanye nendlela yokwahlulela izimangalo ngaphandle kokukhohlisa.
Sebenzisa i-AI ngendlela ekhiqizayo kuyilapho uvikela ubumfihlo, ubheka imiphumela, futhi ulondoloza ukuziphendulela komuntu.
Linganisa izimo zokusetshenziswa kwendawo yokusebenza, sebenzisa abashayeli bezindiza abaphephile, kala inani, futhi uxhumane nezinguquko ngokuzibophezela.
Hlaziya izinhlelo ze-AI ngokusebenzisa amalungelo, ukulingana, ukubusa, ukuphepha, kanye nemiphumela yenzuzo yomphakathi.
Qonda amamodeli olimi, ukubuyisa, abasebenzeli, ukuhlola, izindleko, nokuvikela ukusetshenziswa ngokuklama kwesistimu okusebenzayo.
Amathrekhi esihloko
Gxumela endaweni oyikhathalelayo. Wonke amathrekhi anemihlahlandlela eminingi yesiNgisi esilula.
Umtapo wolwazi ogcwele
117 kwe 1019 imihlahlandlela ebonisiwe. Hlunga ngethrekhi noma sesha ngenhla.
Ukuhlonzwa kwengoma yekhava kuthola lapho okurekhodiwe okubili okunomsindo ohluke kakhulu kuyingoma eyisisekelo efanayo - inguqulo ye-acoustic ebukhoma, i-remix...
Umsindo we-AII-DDSP (I-Differentiable Digital Signal Processing) ihlanganisa amabhulokhi wokwakha we-synthesizer wakudala ngamanethiwekhi e-neural, ukuze ukufunda okujulile kungalawula ama-oscillator...
Umsindo we-AII-VITS imodeli yokuguqula umbhalo ube inkulumo eshintsha umbhalo ngokuqondile uwenze amaza omsindo angahluziwe kusistimu eyodwa eqeqeshiwe, yeqa ipayipi elivamile lezigaba ezimbili.
Umsindo we-AII-FastSpeech ikhiqiza i-spectrogram yenkulumo yonke ngokuhambisana esikhundleni sohlaka olulodwa ngesikhathi, okwenza ukuhlanganisa kusheshe kakhulu futhi kuzinze kakhudlwana.
Umsindo we-AII-NaturalSpeech iwumugqa wocwaningo lwe-Microsoft lwe-TTS oluhloselwe ikhwalithi yenkulumo yezinga lomuntu, enezinguqulo zakamuva ezisebenzisa ukusabalalisa okucashile ukuze kukhiqizwe okucebile, kwemvelo...
Umsindo we-AII-Speech Emotion Recognition (SER) yi-AI ethola isimo somzwelo sesikhulumi - intukuthelo, injabulo, ukudabuka, ukukhungatheka - ezwini lezwi laso, hhayi nje...
Umsindo we-AII-Moshi iwumthombo ovulekile, izwi le-AI lesikhathi sangempela elivela e-Kyutai elikhuluma futhi lilalele ngesikhathi esisodwa - i-duplex egcwele - esikhundleni sokushintshana ngokuqinile.
Umsindo we-AII-Speech-to-Speech Translation (S2ST) ithatha amagama akhulunywayo ngolimi olulodwa futhi ikhiqize amagama akhulunywayo ngolunye — kugcinwa kahle izwi lesikhulumi, iphimbo...
Umsindo we-AII-HiFi-GAN iyi-vocoder ekhiqizayo eshintsha i-mel-spectrogram ibe i-waveform yomsindo oluhlaza cishe ngaso leso sikhathi, ikhiqize inkulumo yekhwalithi yesitudiyo kude…
Umsindo we-AIUkuguqulwa kwe-Grapheme-to-phoneme (G2P) kuhumusha izinhlamvu ezibhaliwe zibe imisindo uhlelo lwenkulumo okufanele luphimisele ngempela.
Umsindo we-AIUkujwayela kombhalo isinyathelo esisekupheleni esiphinda sibhale umbhalo ongahluziwe ube amagama akhulunywa ngokuphelele ngaphambi kokuthi uhlelo lwenkulumo lusho.
Umsindo we-AISoundStream is Google's end-to-end neural audio codec that compresses speech and music to extremely low bitrates while preserving quality.
Check what you learned with a topic quiz, then explore our structured courses or work toward a certificate. Every guide stays free to read.