The Eleven v3 text-to-speech model can be prompted to whisper, laugh, sing, and change tone and accents. It is primarily ...