Pause Of Listener

Updates

20220306-154220

I have a “recorder” script in python that creates 15-second flac files of the sound in my office. Each file is named for the time recording started. For example, 20220306-154220.flac

Then, I have a “transcriber” script that looks for flac files without text transcriptions, sends them to Google for speech-to-text processing, then makes a text file with the same root name as the flac.

20220307-080400

Now the transcriber (soundtotext.py) deletes the sound file if the transcription comes back blank.

I need a male-to-male patch cable to connect other sources to my sound card.

For now, I’d like to figure out a way to save the sound from a Firefox tab, then transcribe that.

20220307-090400

I can monitor the sound from radio scanner channels online via “Broadcastify”, then send them to google for transcription and save the results linked to the audio file.

There’s not much interesting because the audio traffic is mostly inaudible, full of static, or spoken too rapidly for the voice-to-text service to decode.

20220307-091000

I’m putting the Listener project aside for now. I’ve made enough progress to prove that the idea works, and there are other things I’d rather do right now than move this forward. Perhaps I’ll pick it up later. If I do, one thing I’d like to do is find a way to discard silence without ever sending it for transcription. Another thing I’d like to do is try to clean up and slow down the audio before sending it for transcription.