AI-powered audio analysis in the browser. Upload a file, add a Bytez key, get psychological, narrative, technical, and genre analysis plus chat.
- JavaScript 71.3%
- CSS 27.9%
- HTML 0.8%
| css | ||
| js | ||
| .gitignore | ||
| index.html | ||
| LICENSE | ||
| README.md | ||
Audio Forensics Pro
AI-powered audio analysis and deconstruction in the browser. Upload a file, configure a Bytez API key, and get psychological, narrative, technical, cultural, gaming, and genre analysis. Real-time viz placeholders (waterfall, neural, particles, stereo) and power-ups (MIDI, isolation, beat, key, genre) are stubbed for future implementation.
What it does
- Upload / drag & drop — MP3, WAV, FLAC, OGG, AAC, M4A, etc.
- Bytez AI — Save API key (stored in localStorage). Analysis runs on load and fills cards; chat sidebar for follow-up questions.
- Audio window — Choose how much audio is sent per request (10s–60s clip or full file).
- Analysis sections — Psychological, narrative, technical, cultural, gaming, genre. Refresh any section to re-run.
- Player — Loaded file plays in the page; metadata and score placeholders shown.
Run locally
ES modules require a server (no file://):
git clone https://git.hexed.systems/lifelessai/audo.git
cd audo
python3 -m http.server 8080
Open http://localhost:8080. Enter your Bytez key (get one at bytez.com), pick an audio file, click GO!.
Stack
HTML, CSS, vanilla JS (ES modules). Web Audio API, Bytez API for AI. No build step.
License
MIT. See LICENSE.