Open adhikjoshi opened 6 months ago
The model decides the emotion of it's generation based on the emotion of the prompt and the content it will generate. The model currently doesn't support hard coding emotion tag
@jasonppy is it possible to provide separate prompts for the actual words/content and the emotion, or MUST they be bound up together?
How many emotions are supported?
Can we get list of these?
How to trigger specific emotional Audio? Maybe some prompt engineering?