video2doc

Extracts audio tracks from video files using ffmpeg with optimal settings. Supports format conversion, resampling, and channel mixing. Use when extracting audio from video, converting audio formats, preparing audio for speech recognition, or processing media files.

Configures Python projects with uv package manager to use CUDA-enabled PyTorch. Handles the common issue where uv installs CPU-only torch by default. Use when setting up GPU-accelerated machine learning projects, fixing CUDA not available errors, or configuring PyTorch with GPU support in uv-managed projects.

Converts video files (mp4, mkv, webm, avi) to Markdown documents using speech recognition. Extracts audio with ffmpeg, transcribes with OpenAI Whisper, and generates structured Markdown with timestamps. Use when user wants to transcribe video, convert video to text, generate video transcript, or create documentation from video content.

Uses OpenAI Whisper for speech-to-text transcription with GPU acceleration. Handles model selection, language detection, and timestamp extraction. Includes post-processing for hotword correction (fixing misrecognized terms, proper nouns, abbreviations, and technical jargon from ASR output). Use when transcribing audio/video files, extracting speech from media, or performing speech recognition tasks.