ComfyUI Node
FireRedTTS3 Voice Design
Design a brand-new voice from a text description (FireRedTTS3-Instruct).
FireRedTTS3 Voice Design
- firered_model
- audio
- voice_plan
◄instructionA gentle young female voice, speaking a little slowly, with a playful touch.►
◄textHello! This is FireRedTTS3 running natively inside ComfyUI.►
◄languageauto►
◄text_temperature0.70►
◄text_top_p0.80►
◄text_top_k20►
◄text_repetition_penalty1.00►
◄n_timesteps10►
◄inference_cfg1.20►
◄seed42►
◄max_audio_seconds64►
◄do_tntrue►
◄do_splittrue►
◄cross_fade_ms50►
CategoryFireRedTTS3
Inputs (15)
| Name | Type | Default | Description |
|---|---|---|---|
| firered_model | FIREREDTTS3_MODEL | — | |
| instruction | STRING | A gentle young female voice, speaking a little slowly, with a playful touch. | Natural-language voice description (gender, age, timbre, emotion, pace, accent). Any language works; the model card's Chinese examples are also good templates. No reference audio needed. |
| text | STRING | Hello! This is FireRedTTS3 running natively inside ComfyUI. | Text to synthesize with the designed voice. |
| language | COMBO | auto | Language of the text; used for text normalization and sentence splitting only. Voice design synthesis itself only works reliably in Chinese/English (upstream limitation: no language conditioning without a reference clip); other languages produce accented/garbled speech. For other languages, design the voice in English/Chinese, then clone the result with the base model and a language tag. |
| text_temperature | FLOAT | 0.700–2 | Sampling temperature for the model's voice-plan text (Chain-of-Thought). |
| text_top_p | FLOAT | 0.800–1 | Top-p for the voice-plan text. |
| text_top_k | INT | 200–500 | Top-k for the voice-plan text. 0 disables. |
| text_repetition_penalty | FLOAT | 1.001–2 | Repetition penalty for the voice-plan text. 1.0 is the official default. |
| n_timesteps | INT | 101–50 | Flow-matching steps per generated audio patch. 10 is the official default; more is slower with diminishing returns. |
| inference_cfg | FLOAT | 1.200–4 | Classifier-free guidance strength for the flow head. 0 disables CFG. Official defaults: 2.0 for cloning, 1.2 for design/edits. |
| seed | INT | 420–2147483647 | 0 uses the current random state. A positive value is repeatable. |
| max_audio_seconds | FLOAT | 644–160 | Hard cap on generated audio length per sentence (64s is the official maximum). |
| do_tn | BOOLEAN | true | Run text normalization (numbers, dates, units to spoken form). Chinese/English use local wetext; other languages get basic cleaning. |
| do_split | BOOLEAN | true | Split long text into sentences and generate them one by one (cross-faded together). |
| cross_fade_ms | FLOAT | 500–500 | Cross-fade between sentence segments in milliseconds. |
Outputs (2)
| Name | Type | Description |
|---|---|---|
| audio | AUDIO | — |
| voice_plan | STRING | — |