Please add a live music-listening mode to voice chat. I’d like to route an instrument or high-quality audio interface signal into the same input as my voice, then have ChatGPT analyze the music in real time and act as a music teacher.
The feature should distinguish speech from musical audio and provide useful feedback on things such as:
- Pitch and intonation
- Rhythm, timing, tempo, and groove
- Chords, harmony, and key
- Technique and tone
- Mistakes, repetitions, and areas to practise
- Contextual coaching while I play, without requiring me to first record and upload an audio file
Audio fidelity should be preserved enough for meaningful musical analysis, including direct line-level instrument signals. An optional “listen-only” or push-to-talk control would help prevent feedback while playing.
So, here’s a clear draft you can use. Ask for a live music mode in voice chat so you can route an instrument or high-quality interface signal along with your voice, and have ChatGPT analyze the music in real time. It should distinguish speech from musical audio, and give feedback on pitch, rhythm, chords, timing, and technique, like a real-time music analyser or teacher, without needing to upload a recording first.