Hi all. I’m posting this here partially in the hope that it will get back to @theckid, who started me down this rabbit hole with his post on the old forum, and partially because I figured someone else might find it useful.
Anyway, here’s my description from the readme:
A real-time transcription tool for public safety radio feeds (e.g., Broadcastify streams) using OpenAI Whisper (large-v3). Designed for long-running, low-maintenance operation with daily log rotation, robust audio processing, and hallucination filtering.
It should be VERY easy to modify to accept a feed from pretty much any source. It started as trying to clean up the results I was getting from trying to implement theckid’s code and… well, I have a lot of free time on my hands and I may have gotten a little obsessed.
Regardless, here’s the project: GitHub - Nite01007/RadioTranscriber: A Python script for transcribing Broadcastify radio feeds using Whisper AI.
Please feel free to reach out if you have any feedback, or mods, delete this if inappropriate.
Happy new year, y’all!