ComfyUI Node
Generates MOSS Voice from Text
Generates a voice sample based on a text description. Parameters: - instruction: Describe the voice (e.g., 'A young female, soft voice'). - text: The content to be spoken. - audio_temperature: 0.1-2.0. Higher = more variation. - audio_top_p/top_k: Sampling parameters.
Generates MOSS Voice from Text
- moss_vg_model
- audio
◄textHello, this is a test.►
◄instructionA clear, neutral voice for reading.►
◄audio_temperature1.0►
◄audio_top_p0.80►
◄audio_top_k50►
◄audio_repetition_penalty1.0►
◄max_new_tokens2000►
◄text_normalizetrue►
CategoryKaola/MOSS-TTSD
Inputs (9)
| Name | Type | Default | Description |
|---|---|---|---|
| moss_vg_model | MOSS_VOICE_GENERATOR_MODEL | Loaded MOSS-VoiceGenerator model. | |
| text | STRING | Hello, this is a test. | The content to be spoken. |
| instruction | STRING | A clear, neutral voice for reading. | Describe the desired voice characteristics (gender, age, tone, emotion). |
| audio_temperature | FLOAT | 1.00.1–2 | Higher = more variation/drama. |
| audio_top_p | FLOAT | 0.800.1–1 | Nucleus sampling probability. |
| audio_top_k | INT | 501–200 | Top-K sampling. |
| audio_repetition_penalty | FLOAT | 1.01–2 | Penalty for repeating audio tokens. |
| max_new_tokens | INT | 2000100–10000 | Max audio length in tokens. |
| text_normalize | BOOLEAN | true | Normalize text input. |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| audio | AUDIO | — |