ComfyUI Node
ChatterBox TTS (Manual)
A ComfyUI node in audio/latent-astronaut with 10 inputs and 1 output.
ChatterBox TTS (Manual)
- model
- voice_audio
- audio
◄textHello, this is a test of voice cloning.►
◄exaggeration0.50►
◄cfg_weight0.50►
◄temperature0.80►
◄top_p1.00►
◄min_p0.05►
◄repetition_penalty1.20►
◄seed0►
Categoryaudio/latent-astronaut
Inputs (10)
| Name | Type | Default | Description |
|---|---|---|---|
| model | CHATTERBOX_TTS | ChatterBox TTS model from loader | |
| text | STRING | Hello, this is a test of voice cloning. | Text to synthesize into speech |
| voice_audio | AUDIO | Reference audio for voice cloning (6-10 seconds ideal) | |
| exaggeration | FLOAT | 0.500–1 | Emotion/expression intensity (0=neutral, 1=exaggerated) |
| cfg_weight | FLOAT | 0.500–1 | Classifier-free guidance weight (higher=more adherence to text) |
| temperature | FLOAT | 0.800.1–1.5 | Sampling temperature (higher=more varied/random) |
| top_p | FLOAT | 1.000–1 | Nucleus sampling threshold |
| min_p | FLOAT | 0.050–1 | Minimum probability threshold for sampling |
| repetition_penalty | FLOAT | 1.201–2 | Penalty for repeating tokens (1.0=no penalty) |
| seed | INT | 00–18446744073709550000 | Random seed (0=random) |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| audio | AUDIO | — |