Drop a song and get structured data — BPM, key, loudness, duration, format — to paste into ChatGPT, Claude, Gemini, Grok, Perplexity, Suno or Udio so the AI actually understands the track. Everything runs in your browser; your file is never uploaded.
🎧
Drop an audio file or click to choose
It's analyzed on your device — nothing leaves your browser.
MP3 · WAV · OGG · FLAC · AAC · M4A
Analyzing…
Tempo and key are estimated by in-browser DSP — accurate for most music with a clear beat and tonal center, less so for ambient, spoken word, or tempo-changing tracks. Loudness uses the ITU-R BS.1770-4 (LUFS) standard.
What's this for?
AI music tools like Suno and Udio — and chat models like ChatGPT, Claude, Gemini and Grok — can't "listen" to an audio file. If you want to recreate the feel of a reference track, or ask an AI to reason about it, you have to describe it. This tool does that for you: it measures the track's tempo, key, loudness and format locally in your browser and hands you clean JSON (or Markdown) you can paste straight into a chat. No account, no upload. It works with any AI chat or music generator — ChatGPT, Claude, Gemini, Grok, Perplexity, Copilot, DeepSeek, Suno, Udio and more.
FAQ
Is my audio uploaded anywhere?▼
No. Decoding and analysis happen entirely in your browser with the Web Audio API. Your file never leaves your device.
How accurate are the BPM and key?▼
They're estimates from signal analysis — autocorrelation for tempo and a Krumhansl-Schmuckler chromagram for key (the same engines as our BPM Finder and Key Finder). For most music with a clear beat and tonal center they're reliable to within a BPM or two and the right key or its relative; ambient, spoken, or tempo-shifting material is harder. Loudness (LUFS) is a precise ITU-R BS.1770-4 measurement.
How do I use it with Suno or ChatGPT?▼
Pick JSON or Markdown (some chats read Markdown more cleanly), click "Copy for AI chats", paste it into the chat, and ask something like "describe the vibe of a track with these specs" or "write a Suno prompt for something similar." The structured fields give the model concrete, accurate detail to work from instead of guessing.
What fields are included?▼
Duration, sample rate, channels, file format and approximate bitrate; tempo (BPM); musical key, scale, Camelot code and a confidence score; and integrated LUFS loudness plus peak dBFS.
SnipSound