728x90
Step-1 Service-maker ๋ชจ๋ธ ์์ฑ
ngc ๋ฑ๋ก
1. ngc ๊ฐ์
2. nvcr.io์ API ๋ฑ๋ก
docker login nvcr.io
# Username: $oauthtoken
# Password: [ngc API KEY]service-maker๋ก ์ํ๋ ๋ชจ๋ธ ์์ฑ (STT)
1. git clone riva demo
2. ngc pull riva_quickstart
ngc registry resource download-version "nvidia/riva/riva_quickstart:2.8.1"3. riva network set
docker network create riva-speech4. config ํ์ผ ์์
asr_acoustic_model=citrinet_10245. ํ๊ตญ์ด speech model ๋ค์ด๋ก๋
speechtotext_ko_kr_citrinet speechtotext_ko_kr_lm
mkdir servicemaker-dev/models/korean #์์ ์ถ๊ฐ 6. riva-service-maker ๊ตฌ๋& volume ์ถ๊ฐ
bash riva_init.sh
bash riva_start.sh # docker volume (models/korean) ์ถ๊ฐ7. service-maker container์์ model ์์ฑ
#sh ํ์ผ ์์ volume ์์น๋ฅผ customizeํ ๋ค build-deploy ์งํ
scripts/build_deploy/korean_models/riva_asr_citrinet_kr_build.sh
scripts/build_deploy/korean_models/riva_asr_citrinet_kr_deploy.sh๋ฐ์ํ
'๐พ Deep Learning' ์นดํ ๊ณ ๋ฆฌ์ ๋ค๋ฅธ ๊ธ
| [Whisper] Koreanspon Valid (0) | 2023.03.10 |
|---|---|
| [Whisper] (1) - Abstract & Introduction (0) | 2023.03.06 |
| [from numba.np.ufunc import _internalSystemError: initialization of _internal failed without raising an exception] numpy version Error (0) | 2023.02.23 |
| [ASR, ] Deepspeech2 (0) | 2023.02.22 |
| [NVIDIA RIVA] ngc ๋ฑ๋ก (0) | 2023.01.27 |
