Over the last couple of days I have been working on a tool to make Japanese podcast material more comprehensible. I wanted to share the tool here, for people who are interested:
https://github.com/PatrickR1993/youtube-ci-converter
It creates comprehensible input from youtube links or local mp3 audio files by transcribing and translating the audio to english audio. Next, it creates one new audio file where you get EN>JP>EN>JP audio, sentence by sentence. Finally it adds the complete Japanese audio.
You do need OpenAI tokens, but the cost is pretty low, around 0,50 cents for a 15-30 min. podcast.
Let me know what you think!
by SirFunkypants