Docs / source reviewed
Jev Audio Beeper
Beep out flagged words in an audio recording
How this project uses Jev
- Input
- An audio file and a separately supplied word-timestamp transcript
- Jev decides
- Contextual profanity probability and severity for each word
- Code executes
- Code applies a threshold to time intervals; ffmpeg mutes and overlays beeps, with a JSON audit report
Evidence and limitations
The reviewed implementation is an offline file prototype requiring a timestamp sidecar. Streaming ASR and live buffering are proposed extensions. The default threshold is not validated across languages; no runtime test was performed.
This project has not been run independently here. Author-reported results are not independently verified results.