is there a way to generate text from music? (with AI i guess)
i don’t mean to translate the vocals to text-data, more a way of a realtime interpration of the music into text?
could be very dry analytical (“fast paced latin 3/4 ryhthm, with spherical background chords by a synth, hectic male vocals”) or a more expressive way of description which could have own artistic quality. or even funny, commenting on quirky moves in the music.
i ask bc sematics always fall very short in regular audio-reactive nodes. so maybe this could be a way to get more of the vibe and feeling of the music into visuals / music-reactive-content.