whisperx
WhisperX provides advanced speech-to-text capabilities with precise timestamps and speaker identification for efficient transcription.
Install this skill
Security score
The whisperx skill was audited on Jun 4, 2026 and we found 14 security issues across 3 threat categories. Review the findings below before installing.
Categories Tested
Security Issues
Template literal with variable interpolation in command context
| 369 | ```bash |
Access to hidden dotfiles in home directory
| 62 | | **With speakers** | `./scripts/transcribe audio.mp3 --diarize` | Auto-reads `~/.cache/huggingface/token` | |
Access to hidden dotfiles in home directory
| 123 | mkdir -p ~/.cache/huggingface && echo -n "hf_YOUR_TOKEN" > ~/.cache/huggingface/token && chmod 600 ~/.cache/huggingface/token |
Access to hidden dotfiles in home directory
| 133 | - Auto-reads the HF token from `~/.cache/huggingface/token` for diarization |
Access to hidden dotfiles in home directory
| 135 | - First run for a new model downloads it to `~/.cache/huggingface/` (one-time per model) |
Access to hidden dotfiles in home directory
| 171 | # With speaker diarization (auto-reads ~/.cache/huggingface/token) |
Access to hidden dotfiles in home directory
| 259 | --hf-token TOKEN Hugging Face access token (also reads ~/.cache/huggingface/token or HF_TOKEN env) |
External URL reference
| 116 | 1. Create account at [huggingface.co](https://huggingface.co) (if you don't have one) |
External URL reference
| 117 | 2. Go to [huggingface.co/settings/tokens](https://huggingface.co/settings/tokens) and create a **read** access token |
External URL reference
| 119 | - **whisperx ≥3.8.0** (recommended): Accept [pyannote/speaker-diarization-community-1](https://huggingface.co/pyannote/speaker-diarization-community-1) — uses pyannote v4 with better accuracy |
External URL reference
| 120 | - **whisperx <3.8.0**: Accept **both** [pyannote/speaker-diarization-3.1](https://huggingface.co/pyannote/speaker-diarization-3.1) and [pyannote/segmentation-3.0](https://huggingface.co/pyannote/segme |
External URL reference
| 497 | **"CUDA not available"**: Install PyTorch with CUDA (`pip install torch --index-url https://download.pytorch.org/whl/cu121`) |
External URL reference
| 501 | **Diarization 403 error**: You must accept the model agreement(s). For whisperx ≥3.8.0: accept [community-1](https://huggingface.co/pyannote/speaker-diarization-community-1). For earlier versions: acc |
External URL reference
| 528 | - [WhisperX Paper (INTERSPEECH 2023)](https://arxiv.org/abs/2303.00747) |