HAGAHA Audio AI

NVIDIA Riva iyo Hadalka NeMo

NVIDIA Riva waa GPU-dardargelinta SDK ee soo saarista hadalka AI (ASR, TTS, iyo tarjumaada), halka NeMo ay tahay agabka isha furan ee tababarka iyo hagaajinta moodooyinka hoose.

2 daqiiqo akhriMarkii u dambaysay ee la cusbooneysiiyay

Dulmar

Together they let developers build fast, customizable voice applications that run on NVIDIA hardware.

quusid qoto dheer

NeMo (Modules Neural) waa qaab-dhismeedka PyTorch-ka-furan ee NVIDIA si loo dhiso wada-hadalka AI. Waxay soo dirtaa moodallo horay loo sii tababaray oo loogu talagalay aqoonsiga hadalka tooska ah (ASR), qoraal-ku-hadal (TTS), iyo hawlo luqadda dabiiciga ah, oo loo habeeyey sidii ' modules neural' oo dib loo isticmaali karo, waxaad si fiican u hagaajin kartaa xogtaada. Riva waa dhinaca geynta: waxay xirxirtaa moodooyinka la hagaajiyay ee ka dambeeya server-ka gRPC ee qulqulaya, iyadoo la adeegsanayo TensorRT iyo Triton Inference Server si ay ugu dhufato daahitaanka hoose ee cabbirka. Socod-socodka caadiga ah ee shaqada ayaa tabobara ama ku hagaajiya moodal gudaha NeMo, u dhoofiya qaabka Riva, ka dibna u adeega qoraal-waqtiga-dhabta ah ama isku-dhafka. Riva waxay ku taageertaa aqoonsiga qulqulka leh stamps-ka ereyga, codadka TTS neural, hadal-hadal, iyo luqado badan, oo dhamaantood loo habeeyey inay si hufan ugu shaqeeyaan NVIDIA GPUs.

Aragtida Farsamada

Xawaaraha Riva wuxuu ka yimaadaa iskudubarid moodooyin leh TensorRT oo u adeegaya iyada oo loo marayo Triton, taas oo fiyuusa kernels, khusaysa saxnaanta isku dhafan (FP16/INT8), iyo codsiyada isku dhafka ah si firfircooni leh. Moodooyinka ASR sida Conformer-CTC ama Parakeet stream audio oo qaybo yaryar ah iyadoo la ilaalinayo macnaha guud, soo saarista qoraallada qayb ahaan tobanaan ilbiriqsi gudahood. Dhuumaha TTS waxay isku xiraan moodel acoustic ah (tusaale, FastPitch) oo leh cod-yaqaan neural ah (tusaale, HiFi-GAN) si ay u dhaliyaan mowjadaha hirarka si ka dhaqso badan waqtiga dhabta ah ee hal GPU.

Saamaynta Istiraatijiyadeed

Helitaanka iyo gaarsiinta

Waxay wanaajisaa marin u helida iyada oo loo marayo qoraal-qorid, sheeko, iyo is-dhexgalyo cod.

Qiimaha iyo miisaaniyada

Kooxaha warbaahintu waxay ku soo rari karaan codka sifaysan si degdeg ah iyagoo wata miisaaniyado yaryar.

Xawaaraha iyo miisaanka

Nidaamyada u jeedda macmiisha waxay ka baaraandegi karaan isdhexgalka hadalka si weyn.

Mustaqbalka NVIDIA Riva iyo Hadalka NeMo

NVIDIA waxay Riva iyo NeMo u riixaysaa qaabab hadal oo aasaasi ah oo luqado badan leh iyo is dhexgalka adag ee wakiilada LLM ku saleysan ee kaaliyayaasha codka dhamaadka-ilaa-dhamaadka. Filo habaynta qaninimo (kor u qaadista erayada, codadka caadiga ah ee daqiiqadaha xogta), xoojinta deegaanka buuqa badan, iyo geynta ku baahaysa GPU-yada xogta ilaa aaladaha sida Jetson. Maaddaama NeMo ay la socoto moodooyinka wax-soo-saarka, khadka u dhexeeya aqoonsiga hadalka, tarjumaada, iyo sababaynta wada-hadalka ayaa sii wadi doonta inay isku rogto dhuumaha waqtiga-dhabta ah ee midaysan.

Dhaqangelinta Adduunka-dhabta ah

Qoraal-xarunta-waqtiga-dhabta ah iyo wakiilka tooska ah waxay caawiyaan in qoraallada wicitaannada macmiilka ay wataan calaamado heer-erey ah

In la dhiso codad sumadeysan oo TTS ah oo loogu talagalay kaaliyaha farsamada iyadoo la hagaajinayo FastPitch gudaha NeMo dhowr saacadood oo duubis ah

Qoraal toos ah iyo turjumaada hadalka ee shirarka fiidyowga ama baahinta dhacdooyinka NVIDIA GPUs

Hagaajinta qaabka Conformer ASR ee ku saabsan ereyada caafimaadka ama sharciga gaarka ah ee domainka iyadoo la adeegsanayo NeMo, ka dibna u adeega Riva

Khatarta & Dariiqyada Ilaalada

Si xun u isticmaalka codka iyo khataraha is-yeelyeelku way kordhaan marka oggolaanshaha la waayo.

Saxnimadu waxay hoos ugu dhici kartaa lahjadaha, lahjadaha, ama jawiga buuqa badan.

Maqalka synthetic waxaa lagu khaldi karaa hadal dhab ah iyada oo aan si cad loo calaamadin.

Qorshe Hawleedka Dhaqangelinta

1

Hel ogolaansho cad oo ku saabsan qabashada codka, xidhitaanka, iyo dib u isticmaalka

2

Tijaabi tayada ku hadasha kala duwan iyo xaaladaha asalka.

3

Qeex marka bani'aadamku ay tahay inuu dib u eego ama oggolaado wax soo saarka.

4

Ku calaamadee codka synthetic oo xafid diiwaannada la-xisaabtanka.

Sii wad Sahaminta

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 NVIDIA Riva and NeMo Speech quiz

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

Bilow kedis

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

Hagaha xiga

Qiyaasta tayada Hadalka PESQ iyo STOI

Su'aalaha soo noqnoqda

What is NVIDIA Riva and NeMo Speech?

NVIDIA Riva waa GPU-dardargelinta SDK ee soo saarista hadalka AI (ASR, TTS, iyo tarjumaada), halka NeMo ay tahay agabka isha furan ee tababarka iyo hagaajinta moodooyinka hoose. Si wada jir ah waxay u oggolaadeen horumariyayaashu inay si dhakhso leh u dhistaan, codsiyada codka la beddeli karo ee ku shaqeeya qalabka NVIDIA.

Waa maxay farqiga u dhexeeya NeMo iyo Riva?

NeMo waa agabka il-furan ee dhisidda iyo hagaajinta habaynta hadalka iyo moodooyinka luqadda, halka Riva ay xirto oo u adeegto moodooyinkaas isticmaalka wax-soo-saarka hooseeya.

Waa kuwee labada teknoolajiyada ee NVIDIA uu Riva ku tiirsan yahay si uu u gaaro fikradda qarsoon ee hoose?

Riva waxay ku ururisaa moodooyin leh TensorRT oo loogu talagalay fulinta GPU-ga la hagaajiyay waxayna u adeegtaa iyada oo loo marayo Server Inference Server, kaas oo gacanta ku haya dufan firfircoon iyo is-dhaafsi.

Dhuumaha caadiga ah ee Riva TTS, waa maxay doorka cod-bixiyaha sida HiFi-GAN?

Moodeelka acoustic-ka ah sida FastPitch wuxuu saadaaliyaa astaamaha spectrogram ee qoraalka, iyo cod-bixiyaha neural sida HiFi-GAN wuxuu u rogaa astaamahaas qaab mawjada ugu dambeysa ee la maqli karo.

Maxay awood u siinaysaa 'qulqulka' ASR gudaha Riva?

Aqoonsiga socodka qulqulka wuxuu u shaqeeyaa codka qaybo yaryar wuxuuna soo saaraa natiijooyin qayb ah wakhtiga dhabta ah, halkii la sugi lahaa dhammaan odhaahda inuu dhammeeyo.

Keebaa tusaale u ah habaynta NeMo awood u siinaysa moodooyinka hadalka?

NeMo waxay u ogolaataa horumariyeyaasha inay si fiican u habeeyaan moodooyinka horay loo tababbaray xogtooda, tusaale ahaan la qabsiga qaabka ASR si loo aqoonsado ereybixin caafimaad oo khaas ah ama sharci ah.