ComfyUI Node
Qwen TTS Voice Clone Prompt
A ComfyUI node in Vantage/Audio/Qwen3 TTS with 10 inputs and 1 output.
Qwen TTS Voice Clone Prompt
- model
- ref_audio
- voice_prompt
◄ref_text►
◄x_vector_onlyfalse►
◄deviceinherit►
◄attentioninherit►
◄unload_model_after_promptfalse►
◄save_voicefalse►
◄voice_name►
◄overwrite_existingfalse►
CategoryVantage/Audio/Qwen3 TTS
Inputs (10)
| Name | Type | Default | Description |
|---|---|---|---|
| model | QWEN_TTS_MODEL | — | |
| ref_audio | AUDIO | — | |
| ref_textopt | STRING | — | |
| x_vector_onlyopt | BOOLEAN | false | Only use speaker x-vector (no phonetic content) |
| deviceopt | COMBO | inherit | 5 options: inherit, auto, cuda, mps, cpu |
| attentionopt | COMBO | inherit | 6 options: inherit, auto, sage_attn, flash_attn, sdpa, eager |
| unload_model_after_promptopt | BOOLEAN | false | Unload model from memory after building voice prompt |
| save_voiceopt | BOOLEAN | false | Save this voice for reuse |
| voice_nameopt | STRING | — | |
| overwrite_existingopt | BOOLEAN | false | Overwrite existing voice with same name |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| voice_prompt | QWEN_TTS_VOICE_PROMPT | — |