All Speech Recognition Model Downloads | pyVideoTrans Official - Open Source Free Video Translation & Dubbing Software pyvideotrans.com pyvideotrans github github.com/jianchang512/pyvideotrans

The content on this page originates from my open-source project release page. For the latest information, please click GitHub: https://github.com/jianchang512/stt/releases/0.0

faster-whisper Model Downloads, suitable for the stt project and the "pyvideotrans Video Translation & Dubbing" project's faster-whisper mode. For openai-whisper models, please scroll down.

tiny 64MB tiny.en 64MB

base 124MB base.en 124MB

small 415MB small Baidu Netdisk small.en 415MB

medium 1.27G medium.en 1.27G

large-v1 Baidu Netdisk large-v1 huggingface

large-v2 huggingface large-v2 Baidu Netdisk

large-v3 huggingface large-v3 Baidu Netdisk

large-v3-turbo 1.3G

distil-whisper-small.en 282MB

distil-whisper-medium.en 671MB distil-medium Baidu Netdisk

distil-whisper-large-v2 1.27G distil-large-v2 Baidu Netdisk

distil-whisper-large-v3 1.3G distil-whisper-large-v3 Baidu Netdisk

After downloading, extract the archive. Copy the "models--Systran--faster-xx" folder from inside the archive to the models directory. After extraction and copying, the folder list under the models directory should look like this:
Archive contents
Correct folder list under the models directory after placement

openai-whisper Model Downloads, only for the "pyvideotrans Video Translation & Dubbing Software" openai-whisper mode model download and use.

After downloading, place the .pt file into the models folder under the software directory.

tiny.pt model tiny.en.pt model

base.pt model base.en.pt model

large-v3-turbo.pt model

FunASR Chinese Model Download

Baidu Netdisk Download (includes speech recognition, punctuation restoration, and noise reduction models): https://pan.baidu.com/s/1v5wagiid6-K7GX9Pif4reA?pwd=y2ef

Huggingface (Download address outside the firewall): https://huggingface.co/spaces/mortimerme/s4/resolve/main/FunASR-Chinese-models.7z?download=true

After downloading and extracting, you will see three folders: iic, damo, and .__temp. Copy them to the models/hub folder of the video translation software, overwriting if necessary.

cuBLASxx.dll and cudnn Download

If you encounter "cublasxxx.dll not found" or the software crashes after enabling CUDA acceleration, please download the corresponding file. Then copy the dll files inside to the C:/Windows/System32 directory OR to the software root directory (where the exe is located).

Open a command prompt by typing cmd in any folder's address bar, then enter the command nvcc -V to check your current CUDA version.

For CUDA 11.x version, click here to download: https://github.com/jianchang512/stt/releases/download/0.0/cuBLAS.and.cuDNN_CUDA11_win_v4.7z

For CUDA 12.x version, click here to download: https://github.com/jianchang512/stt/releases/download/0.0/cuBLAS.and.cuDNN_CUDA12_win_v1.7z

uvr5 Model Download

Click to download uvr5 model

After downloading, extract the archive. You will get a uvr5_weights folder. Copy this folder to the root directory of the video translation & dubbing software.

ffmpeg.exe Download

If you are on Windows and get a prompt saying the ffmpeg command is not found, you can download the following two files and place them in the software root directory or in the ffmpeg folder under the software root directory.

https://github.com/jianchang512/stt/releases/download/0.0/ffmpeg.exe

https://github.com/jianchang512/stt/releases/download/0.0/ffprobe.exe

faster-whisper Model Downloads, suitable for the stt project and the "pyvideotrans Video Translation & Dubbing" project's faster-whisper mode. For openai-whisper models, please scroll down. ​

openai-whisper Model Downloads, only for the "pyvideotrans Video Translation & Dubbing Software" openai-whisper mode model download and use. ​

FunASR Chinese Model Download ​

cuBLASxx.dll and cudnn Download ​