What a fascinating project! I just got here from a sidebar link that caught my eye while I was reading https://medium.com/mlearning-ai/openai-whisper-is-a-breakthrough-in-speech-recognition-5ac3f4a5bd5d and I can't help wondering how Whisper would improve the quality of your transcriptions...

Having said that, I recently referenced one of his podcasts using https://podscribe.app/, if the transcription work's already done then that might prove a helpful resource :)