Add voice

Creates a new voice from uploaded audio samples. The voice can then be used with text-to-speech and other audio generation endpoints.