Sound Identifier
Find out what can be heard in any recording: speech, music, animals, traffic, alarms and 500 more sounds, with a timeline.
Free, no uploads, no sign-up. Runs 100% in your browser.
How Sound Identifier works
The audio is decoded and resampled to 16 kHz mono, then cut into short windows. An Audio Spectrogram Transformer trained on AudioSet (527 sound classes) labels each window inside a Web Worker, with progress, Cancel, and a CPU fallback.
Input: MP3, WAV, M4A, OGG, MP4. Output: CSV.
How to use Sound Identifier
- Drop an audio or video file.
- Pick how often to check.
- Press Identify and read the timeline.
- Download the timeline as CSV.
Tips
- Nothing is uploaded. Your files stay in this browser tab.
Sound Identifier FAQ
- Are my files uploaded?
- No. This tool runs entirely in your browser tab. Files are read from your disk into memory and the result is generated locally.
- Is it free? Are there limits or watermarks?
- Free and open source (MIT). There are no file-size caps, daily quotas, accounts or watermarks.
- Does it work offline?
- Yes, once the page (and any engine it needs) has loaded, it keeps working without a connection thanks to the service worker cache.
Related tools
- Audio Transcriber: Transcribe speech with on-device Whisper AI and export SRT.
- BPM & Key Finder: Find the tempo and musical key of songs in bulk, with tap tempo to check.
- LUFS Loudness Meter: Measure integrated LUFS, true peak and loudness range, then normalize for Spotify, YouTube or podcasts.
- AI Summarizer: Summarize PDFs, Word files and long text on your device, as paragraphs or key points.