3 ms·
The model is smart enough to not mix them together in the same sample. Different samples would have different voices, unless you put in a voice identifier as in
by backpropaganda 10y ago
The model is smart enough to not mix them together in the same sample. Different samples would have different voices, unless you put in a voice identifier as input to the model. Then you can control which voice to generate with.