Use AI generated images created from prompts transcribed from audio, in this case podcasts. This project is very unfinished, but it works.
convert.py
convert MP3 to mono WAV
transcribe.py
uses Vosk for speach recognition. We feed it an input wav file and get a output-audio-file.txt
generate-art-from-list.py
use Lexica.art to get images from our input text file and output to images-generated.json
Included is a very simple Flask app, see app.py
.