Voice Model V3 - More expressive audio with the support for custom emotions tag like [laugh], [whisper], and many more. Supports 70+ languages.
Voice Model V2 - Faster rendering, supports custom pauses for better pacing and flow control. Supports 29 languages