sing_notes
Generate singing audio directly from phonemes, MIDI note pitches, and per-note durations, so you can create vocal tracks without writing full metadata.
Instructions
用音素、MIDI 音高和每字时长直接生成唱歌,不需要手写完整元数据。
Tool call template
Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {
"phonemes": "<phonemes>",
"note_pitches": "<note_pitches>",
"durations_ms": "<durations_ms>"
},
"name": "sing_notes"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| text | No | ||
| device | No | auto | |
| language | No | Mandarin | |
| phonemes | Yes | ||
| auto_shift | No | ||
| pitch_shift | No | ||
| durations_ms | Yes | ||
| note_pitches | Yes | ||
| output_format | No | mp3 | |
| prompt_wav_path | No | ||
| normalize_loudness | No | ||
| prompt_metadata_path | No |