Text & data
Transcribe audio to text
Upload an audio file and get a text transcript you can copy or download — the speech model may download once and cache on your device.
🔒 Your file stays on this device — processing runs in your browser.
Save and continue on the desktop
Download or copy your result here first. Continue in nopa to save to Files or use more features.
- Transcribe audio to text locally — no upload
- Results saved to Files automatically
- Every tool in one place — no new tabs
After you download — continue on the desktop for Files or the full app.
What Local Transcriber can do
Speech to text in your browser
Upload MP3, WAV, or M4A and get a transcript you can copy or download — no signup. The speech model may download once, then runs on your device.
Honest about limits
Works best on desktop Chrome or Edge with short audio first. Phones may be slow or unsupported depending on the browser.
Supported formats
MP3, WAV, M4A, OGG, and WebM audio files. Video files are not supported yet — extract audio first.
Your files stay on your device
While you work, files are handled in your browser on this device — not sent off for processing.
Related apps
Questions
Is my audio uploaded?
No. Transcription runs locally in your browser when supported. The speech model may download once from a public model hub and cache on your device.
What languages are supported?
The default multilingual model supports many languages including English, Russian, Spanish, and German. Pick the audio language or use auto-detect.
Will it work on my phone?
Some phones support local transcription but it may be slow. For best results, use Chrome or Edge on a desktop with short audio first.
What license covers the speech model?
nopa uses Xenova Whisper models (OpenAI Whisper, MIT) via Transformers.js (Apache-2.0). See /licenses for details.