New Media Lab Whisper Transcriber

Upload audio/video and get SRT/VTT/TXT.

Processing Mode

⚠️
Cloud Mode:
  • Model: Whisper large-v3-turbo (faster & more accurate than base Whisper)
  • File size limit: 50MB maximum (all formats supported including MP3, M4A, WAV, MP4)
  • Files sent to Cloudflare Workers AI (not stored, but use Local Mode for sensitive content)
  • Speaker diarization not available (Local Mode only)
Audio Transcription Upload audio or video files to generate automated transcriptions.
Important: All AI-generated transcriptions should be reviewed manually for accuracy before use.
📁
Drag & drop a file here, or click to browse
Supports MP3, M4A, WAV, MP4, and other audio/video formats
⚠️ Maximum file size: 50MB (Cloud Mode). Large files may take several minutes to process.
⏱️ Processing times longer than 2 minutes? Use our direct access link to avoid timeouts on very long files.
Advanced Settings
Automatically identify different speakers in the recording (e.g., interviewer vs. interviewee)
← Back to Tools
Progress
Ready
0%

Speaker Identification