Vibe TogetherVVibe Together
Aug 16

More expressive vocal synthesis

We are exploring DDSP and other vocal-synthesis approaches to make Vocal Nodes and Songbook performances feel more connected and expressive across syllables, pitches, dynamics, and longer phrases. The review covers model and voice licensing, performer consent, pronunciation control, client versus server execution, latency, constrained-device fallbacks, and parity between live playback and exported audio. The goal is better musical vocals without hiding when a voice or rendering mode has different capabilities.
ReviewingReviewing