| gradio | |
| torch | |
| torchvision | |
| torchaudio | |
| spaces | |
| imageio-ffmpeg | |
| # SoniTranslate Core Dependencies | |
| git+https://github.com/m-bain/whisperX.git | |
| pyannote.audio>=3.3.2 | |
| fairseq | |
| yt-dlp | |
| pysrt | |
| pydub | |
| faster-whisper | |
| audiostretchy | |
| # Translation and TTS | |
| google-generativeai | |
| openai | |
| edge-tts | |
| piper-tts==1.2.0 | |
| TTS==0.21.1 | |
| # Other utilities | |
| # Important: numpy must be <2 for audio libraries to work | |
| numpy<2 | |
| soundfile | |
| librosa | |
| onnxruntime-gpu | |
| tqdm | |
| demucs | |
| python-multipart | |
| tenacity | |
| youtube-transcript-api | |
| ffmpeg-python |