MattCowski
commented
3 weeks ago Any luck? I found this: https://www.npmjs.com/package/twilio-media-stream-save-audio-file Which then could be sent again to openai completions for a transcription. But it would be better if we could get both the audio and transcriptions directly from the stream. Any advice on how the code should look?
I'm particularly interested in the possibility of retrieving the conversation's audio, whether they are in one or separate files. I'd be really grateful if someone knows how to achieve this.