Dia TTSDia TTS

Dia TTS: Natural-Sounding AI-Powered Speech Generator

0 Credits

Optional audio file (.wav/.mp3/.flac) for voice cloning. The model will attempt to mimic this voice style.

Drop Audio Here

- or -

Click to Upload

Generation Parameters

Controls randomness. Higher values (e.g., 1.3-2.0) increase variety/expressiveness; lower values make output more consistent. Default: 1.3

Adjusts playback speed. Values below 1.0 slow down the audio; 1.0 is original speed. Default: 0.94

Controls diversity/creativity. Lower values focus on more likely outputs. Default: 0.95

Controls how closely the audio follows the text. Higher values (e.g., 3-5+) follow text more strictly. Default: 4

Generated audio will appear here