1. Home
  3. AI
  5. AI Sound Classifier

AI Sound Classifier

Detect speech, music, dogs, sirens, rain, vehicles, applause and 500+ sounds with a timeline, on your device.

AI Private · runs on your device Free
Runs locally in your browser AI processing happens on your device. Your files are not uploaded for inference.
AI model
…
Download size
… ·
Device
Checking browser support…

    What sounds are in the recording?

    The Sound Classifier uses YAMNet, trained on Google's AudioSet, to score 521 everyday sound classes in windows of about one second. Speech, music, laughter, applause, dog barks, cat meows, sirens, horns, rain, wind, engines and much more are recognised.

    Two ways to use it

    • File analysis: upload an audio or video file to get the most prominent sounds and a timeline of what was heard when
    • Live microphone: classify the sounds around you in real time

    Use cases

    Tagging audio archives, finding specific sounds in video content, environmental sound research, accessibility projects and AI teaching. All analysis happens in your browser; your recordings aren't uploaded.

    How to use AI Sound Classifier

    1. Upload an audio or video file or start the microphone.
    2. The model analyses the audio in one-second windows.
    3. Review the dominant sounds and the timeline.
    4. Use the results as you need.

    Why use this tool?

    Discover what's in long recordings quickly without listening from start to finish.

    FAQ

    Why does it show several sounds at once?

    Real recordings are often mixed (speech over music, traffic in rain). The model scores each class independently, so overlapping sounds can all score high.

    Does it transcribe spoken words?

    No, it only recognises the type of sound. Use the AI Video Subtitle Generator to transcribe speech.

    Is the microphone audio recorded?

    No, the stream is analysed on the fly and not saved.

    Related tools

    Private

    BPM Finder (Tempo Detector)

    Estimate a song's tempo (BPM) with beat markers and half/double-tempo alternatives.

    Open tool
    Private

    Song Key Finder

    Estimate a song's key (e.g. A minor or C major) with confidence, relative key and Camelot code.

    Open tool
    Private

    AI Voice Activity Detector (VAD)

    Find where people speak and where it's silent, with timestamps, totals and an exportable segment list.

    Open tool
    Private

    AI Voice Noise Remover

    Remove fan, hum, traffic and room noise from voice recordings with RNNoise; compare and download a clean WAV.

    Open tool