ComfyUI Node
Qwen3 TTS 语音合成（2 合 1）
A ComfyUI node in Seedance with 9 inputs and 5 outputs.
Qwen3 TTS 语音合成（2 合 1）
- api_config
- audio
- audio_url
- audio_path
- task_id
- response
◄modelqwen3-tts-flash►
◄prompt►
◄voiceCherry►
◄language_typeChinese►
◄instructions►
◄optimize_instructionstrue►
◄skip_errorfalse►
◄seed0►
CategorySeedance
Inputs (9)
| Name | Type | Default | Description |
|---|---|---|---|
| model | COMBO | qwen3-tts-flash | Flash reads text directly; Instruct Flash also accepts delivery instructions. | Flash 直接朗读，Instruct Flash 可接受表达指令。 |
| prompt | STRING | Text to synthesize. | 需要合成的文本。 | |
| voice | STRING | Cherry | Qwen3 TTS system or custom voice ID. | Qwen3 TTS 系统音色或自定义音色 ID。 |
| language_type | COMBO | Chinese | Language matching the input text. | 与输入文本匹配的语言。 |
| instructions | STRING | Instruct Flash only: natural-language speaking style, in Chinese or English. | 仅 Instruct Flash：中文或英文表达风格指令。 | |
| optimize_instructions | BOOLEAN | true | Instruct Flash only: refine non-empty instructions upstream. | 仅 Instruct Flash：优化非空表达指令。 |
| api_configopt | SEEDANCE_CONFIG | Connect Seedance API Config; otherwise SEEDANCE_API_KEY is used. | |
| skip_erroropt | BOOLEAN | false | On failure return one second of silence instead of stopping the workflow. | 失败时输出 1 秒静音。 |
| seedopt | INT | 00–18446744073709550000 | ComfyUI cache seed. Fixed reuses the cached result while all other inputs stay unchanged; randomize/increment/decrement starts a new execution. This value is not sent to models without documented seed support. | ComfyUI 缓存种子；Fixed 在其他输入不变时复用缓存，随机、递增或递减会触发新任务。未声明支持 seed 的模型不会收到此参数。 |
Outputs (5)
| Name | Type | Description |
|---|---|---|
| audio | AUDIO | — |
| audio_url | STRING | — |
| audio_path | STRING | — |
| task_id | STRING | — |
| response | STRING | — |