Video Transcript Tool
Upload a video or audio file and generate a transcript directly in your browser with a local Whisper model.
Supported examples: mp4, mov, mp3, wav, m4a, webm. For better reliability, keep uploads under 25MB.
First run downloads the model in your browser cache. Larger models are slower but usually more accurate.
For multilingual model only. Use ISO-639-1 (en, es, fr, de). Leave blank for auto-detect.
Note: no speech model can guarantee a mathematically perfect transcript on every recording. For best results, use clear audio and minimal background noise.