Ubuyobozi bwa Audio AI

NVIDIA Riva na Ijambo rya NeMo

NVIDIA Riva ni GPU yihuta ya SDK yo kuvuga umusaruro AI (ASR, TTS, hamwe nubuhinduzi), naho NeMo nigitabo gifungura isoko yo guhugura no guhuza neza imiterere yibanze.

Umva iki gitaboAI-byavuzwe · ~2 min

Kuriyi page2 min soma
  1. Incamake
  2. Kwibira cyane
  3. Gushyira mu bikorwa Isi
  4. Ejo hazaza ha NVIDIA Riva na NeMo Imvugo
  5. Kubishyira mubikorwa
  6. Komeza Ubushakashatsi
  7. Ibibazo bikunze kubazwa

Incamake

Hamwe na hamwe bareka abitezimbere bakubaka byihuse, byemewe amajwi akoreshwa kuri NVIDIA.

Kwibira cyane

NeMo (Neural Modules) ni NVIDIA ifungura-isoko ya PyTorch yo kubaka AI iganira. Kohereza ibyitegererezo byateganijwe kugirango bimenyekanishe mu buryo bwikora (ASR), inyandiko-ku-mvugo (TTS), n'imirimo y'ururimi karemano, byateguwe nka 'modules modules' byongeye gukoreshwa ushobora guhuza neza amakuru yawe bwite. Riva ni uruhande rwoherejwe: ipakira moderi nziza cyane inyuma ya seriveri ya gRPC, ikoresheje TensorRT hamwe na Triton Inference Server kugirango ikubite ubukererwe buke mubipimo. Ubusanzwe akazi gahugura cyangwa gahuza icyitegererezo muri NeMo, kyohereza muburyo bwa Riva, hanyuma ikagikora mugihe cyo kwandukura cyangwa kugihe. Riva ishyigikira kumenyekanisha hamwe nijambo-urwego rwigihe, amajwi ya TTS, amajwi ya disikuru, hamwe nindimi nyinshi, byose byahujwe kugirango bikore neza kuri NVIDIA GPUs.

Gushyira mu bikorwa Isi

Igihe nyacyo cyo guhamagarira-kwandikirana hamwe na agent nzima ifasha yanditseho abakiriya bahamagara hamwe nijambo-urwego rwigihe

Kubaka amajwi yihariye ya TTS kumufasha wungirije muburyo bwiza bwo gutunganya FastPitch muri NeMo kumasaha make yafashwe

Kwandika neza no gusobanura imvugo kubiganiro bya videwo cyangwa ibyabaye kuri NVIDIA GPUs

Kuringaniza neza moderi ya ASR ikora kumurongo wihariye wubuvuzi cyangwa amategeko yemewe ukoresheje NeMo, hanyuma ukayikorera ukoresheje Riva

Ejo hazaza ha NVIDIA Riva na NeMo Imvugo

NVIDIA irasunika Riva na NeMo kugana imvugo nini, indimi nyinshi zerekana imvugo no kwishyira hamwe hamwe na LLM ishingiye kubakozi bafasha amajwi arangira. Tegereza ibintu bikize cyane (kuzamura ijambo, amajwi yihariye kuva muminota yamakuru), urusaku rwiza-ibidukikije bikomeye, hamwe no kohereza amakuru hagati ya GPUs kugeza kubikoresho nka Jetson. Mugihe NeMo igenda ihindagurika hamwe nicyitegererezo kibyara umusaruro, umurongo uri hagati yo kumenyekanisha imvugo, guhindura, no gutekereza kubitekerezo bizakomeza guhurizwa mumiyoboro ihuriweho nigihe nyacyo.

Kubishyira mubikorwa

Impamvu ari ngombwa

  • Kugera no kugera. Itezimbere kugerwaho binyuze mu kwandukura, kuvuga, no guhuza amajwi.
  • Igiciro na bije. Amatsinda yibitangazamakuru arashobora kohereza amajwi yihuse hamwe na bije nto.
  • Umuvuduko n'igipimo. Sisitemu ireba abakiriya irashobora gutunganya imikoranire ivugwa murwego runini.

Witondere

  • Gukoresha nabi amajwi no kwigira ibyago byiyongera mugihe uruhushya rubuze.
  • Ukuri kurashobora kugabanuka hejuru yimvugo, imvugo, cyangwa urusaku rwibidukikije.
  • Amajwi yubukorikori arashobora kwibeshya kumvugo yukuri nta kirango gisobanutse.

Intambwe ikurikira

  1. Shaka uruhushya rusobanutse rwo gufata amajwi, gukoroniza, no gukoresha.
  2. Ikizamini cyiza mubiganiro bitandukanye hamwe nuburyo bwimbere.
  3. Sobanura igihe umuntu agomba gusuzuma cyangwa kwemeza ibisubizo.
  4. Andika amajwi yubukorikori kandi ugumane inyandiko zerekana kubazwa.

Komeza Ubushakashatsi

Ibaruwa y'amakuru y'ubuntu

Ibuka amakuru ya buri munsi ya AI

Inkuru eshatu zizewe za AI buri gitondo cy'iminsi y'akazi, zanditse mu rurimi rworoshye. Ubuntu iteka ryose, nta kwamamaza.

Imeri imwe buri munsi y'akazi. Kwiyimura mu buryo bumwe. Ntitwigera tugurisha cyangwa ngo dusangire aderesi yawe.

Isuzume

Take the NVIDIA Riva and NeMo Speech quiz

Ibisubizo ako kanya kuri buri kibazo, n'icyemezo gishobora gusangirwa gifite ID yagenzurwa umaze gutsinda isomo.

Tangira ikibazo

Fasha uburezi bwa AI bw'ubuntu. AI Understanding ni umuryango utegamiye kuri leta wa 501(c)(3) — nta kwamamaza, nta kwishyura, iteka ryose. Tanga impano

Ibibazo bikunze kubazwa

Niki NVIDIA Riva na NeMo Imvugo?

NVIDIA Riva ni GPU yihuta ya SDK yo kuvuga umusaruro AI (ASR, TTS, hamwe nubuhinduzi), naho NeMo nigitabo gifungura isoko yo guhugura no guhuza neza imiterere yibanze. Hamwe na hamwe bareka abitezimbere bakubaka byihuse, byemewe amajwi akoreshwa kubikoresho bya NVIDIA.

Ni irihe tandukaniro ryibanze riri hagati ya NeMo na Riva?

NeMo nigikoresho gifungura ibikoresho byubaka no gutunganya neza imvugo nicyitegererezo cyururimi, mugihe Riva ipakira kandi ikora izo moderi zo gukoresha umusaruro muke.

Which NVIDIA components does Riva documentation identify for optimized model execution and inference serving?

TensorRT and Triton have distinct execution and serving roles; particular batching and latency behavior depends on the deployed pipeline.

Muburyo busanzwe bwa Riva TTS, ni uruhe ruhare rwijwi nka HiFi-GAN?

Moderi ya acoustic nka FastPitch ihanura ibiranga spekrogramu uhereye kumyandiko, hamwe nijwi ryumvikana nka HiFi-GAN rihindura ibyo bintu muburyo bwanyuma bwumvikana.

Niki 'gutembera' ASR muri Riva itanga?

Kumenyekanisha gutembera gutunganya amajwi mubice bito kandi bigatanga ibisubizo byigice mugihe nyacyo, aho gutegereza ko amagambo yose arangira.

Niyihe ngero yo kwihitiramo NeMo itanga uburyo bwo kuvuga?

NeMo ireka abitegura gukora neza-moderi yabigenewe ku makuru yabo bwite, urugero guhuza imiterere ya ASR kugirango bamenye imvugo yihariye yubuvuzi cyangwa amategeko.