Request ability to support basic TTS markup for better control over pauses, emphasis, pace, volume, etc.
#18
by
Booly7
- opened
This is an amazing model and the clean interface provided makes it easy to start using it right away. First thing I noticed is that it does not consistently follow punctuation, especially with longer text prompts or when the source audio provides less variety. Adding support for some basic TTS markup like specific pause indicators, volume, emphasis, emotion, or relative speed tags, would be super helpful. Thanks!