There are currently only 3 voice models available — we need more options, as seen in other languages.

Models with faster or slower speech, and even some with regional accents, would be very helpful.