Introduction to Producer Music Generation
Welcome to the Producer Music Generation service documentation!
Producer is a powerful AI music generation service that allows you to create, edit, and transform music with advanced capabilities including negative prompts, time-range control, and various editing operations.
Features
Producer AI empowers you with advanced music generation features:
- Create Music: Generate original music tracks with style prompts, negative prompts, and time-range control
- Extend Music: Continue existing tracks seamlessly
- Cover Music: Transform music into different styles while preserving structure
- Replace Section: Replace specific portions of a track
- Swap Vocals: Replace vocals while keeping the instrumental
- Swap Sound: Replace instrumental while keeping vocals
- Music Variation: Create variations of existing tracks
Available Models
| Model | Description |
|---|
FUZZ-2.0 Pro | Latest flagship model with best quality |
FUZZ-2.0 | Standard 2.0 model |
FUZZ-2.0 Raw | Raw output model |
FUZZ-1.1 Pro | Professional 1.1 model |
FUZZ-1.1 | Standard 1.1 model |
FUZZ-1.0 Pro | Professional 1.0 model |
FUZZ-1.0 | Standard 1.0 model |
FUZZ-0.8 | Legacy model |
How to Use Producer
Create a Task
Call the appropriate endpoint (e.g., /producer/audios) to create a generation task.
Get Task ID
Receive a taskId from the response.
Poll for Results
Use /producer/tasks with your taskId to poll until status is completed.
Download Results
Use /producer/download or /producer/download-video to get your audio in desired format.
API Endpoints
| Endpoint | Method | Description |
|---|
/producer/audios | POST | Main audio generation/editing endpoint |
/producer/tasks | POST | Query task status and results |
/producer/stems | POST | Get stem separation |
/producer/upload | POST | Upload audio file |
/producer/download | POST | Download audio in various formats |
/producer/download-video | POST | Generate and download video |
Action Examples
The /producer/audios endpoint uses the action parameter to determine the operation. Below are complete examples for each action type.
1. create_music - Create New Music
Create original music with style prompts, negative prompts, and time-range control.
Negative Prompts: Use negative strength values (e.g., -0.5) to avoid certain styles or characteristics.Time Range: Use timeStart and timeEnd in sounds array to apply styles to specific time ranges.
{
"action": "create_music",
"model": "FUZZ-2.0 Pro",
"title": "夏天",
"seed": 349732,
"sounds": [
{ "text": "emotional pop with gentle piano, warm synths, and a catchy beat", "strength": 0.5 },
{ "text": "吐字模糊", "strength": -0.5 },
{ "text": "跑调", "strength": -0.5 },
{ "text": "卡断", "strength": -0.5, "timeStart": 0 }
],
"lyrics": "【主歌 A】\n窗外的雨轻轻落下\n打湿了旧日的牵挂\n桌角还放着那杯冷茶\n像我们说过的悄悄话\n\n翻开的相册有些泛黄\n笑容却依然在发光\n那些走过的路和远方\n都藏在心底最柔软的地方\n\n【副歌】\n时光里的我们\n像风吹过黄昏\n有过天真,有过认真\n也有说不出的心疼\n\n回忆里的我们\n握紧彼此的手纹\n就算世界变得陌生\n爱依然温热完整\n\n【主歌 B】\n城市的灯慢慢亮起\n像星空落满大地\n我们曾许下小小秘密\n说要一起走到结局\n\n人海中偶尔会迷路\n却从未真的孤独\n因为心里总有个你\n替我照亮未来的旅途\n\n【副歌】\n时光里的我们\n像风吹过黄昏\n有过天真,有过认真\n也有说不出的心疼\n\n回忆里的我们\n握紧彼此的手纹\n就算世界变得陌生\n爱依然温热完整\n\n【桥段】\n也许岁月会改变模样\n让誓言不再闪亮\n但只要心还记得方向\n梦就不会被遗忘\n\n【尾副歌】\n时光里的我们\n穿越四季晨昏\n走过离分,走过沉沦\n依然愿意再拥抱一次青春\n\n未来里的我们\n会更懂得温存\n就算明天遥远纷纷\n爱依然为彼此永恒\n\n【尾声】\n愿多年以后回头\n还能微笑着说——\n这是属于我们的温柔",
"lyricsStrength": 0.5,
"weirdness": 0.5,
"makeInstrumental": false
}
2. extend_music - Extend Existing Music
Continue an existing track from a specified position.
{
"action": "extend_music",
"model": "FUZZ-2.0 Pro",
"audioId": "32981605-4853-450a-8d0b-ecdca783b51a",
"sounds": [
{ "text": "emotional pop with gentle piano, warm synths, and a catchy beat", "strength": 0.5 }
],
"lyrics": "清晨的风轻轻吹过窗口\n把梦里的心事悄悄带走\n街角的灯还舍不得熄灭\n像我们曾紧握的温柔\n\n这一刻世界变得很轻柔\n时间在指尖慢慢停留\n只要你还在我的左右\n平凡日子也闪着微光的节奏",
"title": "",
"seed": 349732,
"lyricsStrength": 0.5,
"weirdness": 0.5,
"continueAt": 212,
"extendDuration": 30,
"makeInstrumental": false
}
| Parameter | Description |
|---|
audioId | The source audio ID to extend |
continueAt | Start position in seconds |
extendDuration | Duration to extend in seconds |
3. replace_music - Replace Section
Replace a specific section of an existing track with new content.
{
"action": "replace_music",
"model": "FUZZ-2.0 Pro",
"audioId": "32981605-4853-450a-8d0b-ecdca783b51a",
"sounds": [
{ "text": "emotional pop with gentle piano, warm synths, and a catchy beat", "strength": 0.5 }
],
"lyrics": "夜空的风轻轻落下\n吹散了往事的牵挂\n窗边仍留着那盏灯花\n像心中藏着的悄悄话",
"title": "replace",
"seed": 349732,
"lyricsStrength": 0.5,
"weirdness": 0.5,
"replaceStart": 15,
"replaceEnd": 28,
"makeInstrumental": false
}
| Parameter | Description |
|---|
audioId | The source audio ID |
replaceStart | Start time in seconds of section to replace |
replaceEnd | End time in seconds of section to replace |
4. cover_music - Cover Music
Transform music into a different style while preserving structure.
{
"action": "cover_music",
"model": "FUZZ-2.0 Pro",
"audioId": "32981605-4853-450a-8d0b-ecdca783b51a",
"sounds": [
{ "text": "high-energy upbeat pop with driving electronic beats, punchy bass, dynamic synth leads, and powerful rhythmic drops, designed for an energetic and explosive performance", "strength": 0.5 }
],
"lyrics": "【主歌 A】\n窗外的雨轻轻落下\n打湿了旧日的牵挂\n桌角还放着那杯冷茶\n像我们说过的悄悄话\n\n翻开的相册有些泛黄\n笑容却依然在发光\n那些走过的路和远方\n都藏在心底最柔软的地方\n\n【副歌】\n时光里的我们\n像风吹过黄昏\n有过天真,有过认真\n也有说不出的心疼\n\n回忆里的我们\n握紧彼此的手纹\n就算世界变得陌生\n爱依然温热完整\n\n【主歌 B】\n城市的灯慢慢亮起\n像星空落满大地\n我们曾许下小小秘密\n说要一起走到结局\n\n人海中偶尔会迷路\n却从未真的孤独\n因为心里总有个你\n替我照亮未来的旅途\n\n【副歌】\n时光里的我们\n像风吹过黄昏\n有过天真,有过认真\n也有说不出的心疼\n\n回忆里的我们\n握紧彼此的手纹\n就算世界变得陌生\n爱依然温热完整\n\n【桥段】\n也许岁月会改变模样\n让誓言不再闪亮\n但只要心还记得方向\n梦就不会被遗忘\n\n【尾副歌】\n时光里的我们\n穿越四季晨昏\n走过离分,走过沉沦\n依然愿意再拥抱一次青春\n\n未来里的我们\n会更懂得温存\n就算明天遥远纷纷\n爱依然为彼此永恒\n\n【尾声】\n愿多年以后回头\n还能微笑着说——\n这是属于我们的温柔",
"title": "cover",
"seed": 349732,
"lyricsStrength": 0.5,
"weirdness": 0.5,
"coverStrength": 0.5,
"coverArtUrl": "https://dkkj.s3.ap-east-1.amazonaws.com/tmp/123.png",
"makeInstrumental": false
}
| Parameter | Description |
|---|
audioId | The source audio ID to cover |
coverStrength | How much to transform (0-1) |
coverArtUrl | Optional cover art image URL |
5. swap_music_sound - Swap Instrumental
Replace the instrumental/backing track while keeping vocals.
{
"action": "swap_music_sound",
"model": "FUZZ-2.0 Pro",
"audioId": "32981605-4853-450a-8d0b-ecdca783b51a",
"sounds": [
{ "text": "high-energy classic disco style with bright funky basslines, groovy rhythm guitar, vibrant brass sections, punchy four-on-the-floor beats, retro string stabs, and lively dancefloor atmosphere, completely transforming the original arrangement into an unmistakable disco production", "strength": 0.9 }
],
"lyrics": "【主歌 A】\n窗外的雨轻轻落下\n打湿了旧日的牵挂\n桌角还放着那杯冷茶\n像我们说过的悄悄话\n\n翻开的相册有些泛黄\n笑容却依然在发光\n那些走过的路和远方\n都藏在心底最柔软的地方\n\n【副歌】\n时光里的我们\n像风吹过黄昏\n有过天真,有过认真\n也有说不出的心疼\n\n回忆里的我们\n握紧彼此的手纹\n就算世界变得陌生\n爱依然温热完整\n\n【主歌 B】\n城市的灯慢慢亮起\n像星空落满大地\n我们曾许下小小秘密\n说要一起走到结局\n\n人海中偶尔会迷路\n却从未真的孤独\n因为心里总有个你\n替我照亮未来的旅途\n\n【副歌】\n时光里的我们\n像风吹过黄昏\n有过天真,有过认真\n也有说不出的心疼\n\n回忆里的我们\n握紧彼此的手纹\n就算世界变得陌生\n爱依然温热完整\n\n【桥段】\n也许岁月会改变模样\n让誓言不再闪亮\n但只要心还记得方向\n梦就不会被遗忘\n\n【尾副歌】\n时光里的我们\n穿越四季晨昏\n走过离分,走过沉沦\n依然愿意再拥抱一次青春\n\n未来里的我们\n会更懂得温存\n就算明天遥远纷纷\n爱依然为彼此永恒\n\n【尾声】\n愿多年以后回头\n还能微笑着说——\n这是属于我们的温柔"
}
Use higher strength values (e.g., 0.9) for more dramatic instrumental transformation.
6. swap_music_vocals - Swap Vocals
Replace vocals while keeping the instrumental track.
{
"action": "swap_music_vocals",
"model": "FUZZ-2.0 Pro",
"audioId": "32981605-4853-450a-8d0b-ecdca783b51a",
"sounds": [
{ "text": "dramatic vocal transformation into a clearly male voice, deep and warm baritone tone, rich emotional delivery, expressive and powerful singing style, completely different timbre and gender from the original performance, with confident phrasing and distinctive character", "strength": 1 }
],
"lyrics": "【主歌 A】\n窗外的雨轻轻落下\n打湿了旧日的牵挂\n桌角还放着那杯冷茶\n像我们说过的悄悄话\n\n翻开的相册有些泛黄\n笑容却依然在发光\n那些走过的路和远方\n都藏在心底最柔软的地方\n\n【副歌】\n时光里的我们\n像风吹过黄昏\n有过天真,有过认真\n也有说不出的心疼\n\n回忆里的我们\n握紧彼此的手纹\n就算世界变得陌生\n爱依然温热完整\n\n【主歌 B】\n城市的灯慢慢亮起\n像星空落满大地\n我们曾许下小小秘密\n说要一起走到结局\n\n人海中偶尔会迷路\n却从未真的孤独\n因为心里总有个你\n替我照亮未来的旅途\n\n【副歌】\n时光里的我们\n像风吹过黄昏\n有过天真,有过认真\n也有说不出的心疼\n\n回忆里的我们\n握紧彼此的手纹\n就算世界变得陌生\n爱依然温热完整\n\n【桥段】\n也许岁月会改变模样\n让誓言不再闪亮\n但只要心还记得方向\n梦就不会被遗忘\n\n【尾副歌】\n时光里的我们\n穿越四季晨昏\n走过离分,走过沉沦\n依然愿意再拥抱一次青春\n\n未来里的我们\n会更懂得温存\n就算明天遥远纷纷\n爱依然为彼此永恒\n\n【尾声】\n愿多年以后回头\n还能微笑着说——\n这是属于我们的温柔"
}
Describe the desired vocal characteristics in detail (gender, tone, style) for best results.
7. music_variation - Create Variations
Generate variations of an existing track.
{
"action": "music_variation",
"model": "FUZZ-2.0 Pro",
"audioId": "32981605-4853-450a-8d0b-ecdca783b51a",
"sounds": [
{ "text": "", "strength": 0.5 }
],
"lyrics": "【主歌 A】\n窗外的雨轻轻落下\n打湿了旧日的牵挂\n桌角还放着那杯冷茶\n像我们说过的悄悄话\n\n翻开的相册有些泛黄\n笑容却依然在发光\n那些走过的路和远方\n都藏在心底最柔软的地方\n\n【副歌】\n时光里的我们\n像风吹过黄昏\n有过天真,有过认真\n也有说不出的心疼\n\n回忆里的我们\n握紧彼此的手纹\n就算世界变得陌生\n爱依然温热完整\n\n【主歌 B】\n城市的灯慢慢亮起\n像星空落满大地\n我们曾许下小小秘密\n说要一起走到结局\n\n人海中偶尔会迷路\n却从未真的孤独\n因为心里总有个你\n替我照亮未来的旅途\n\n【副歌】\n时光里的我们\n像风吹过黄昏\n有过天真,有过认真\n也有说不出的心疼\n\n回忆里的我们\n握紧彼此的手纹\n就算世界变得陌生\n爱依然温热完整\n\n【桥段】\n也许岁月会改变模样\n让誓言不再闪亮\n但只要心还记得方向\n梦就不会被遗忘\n\n【尾副歌】\n时光里的我们\n穿越四季晨昏\n走过离分,走过沉沦\n依然愿意再拥抱一次青春\n\n未来里的我们\n会更懂得温存\n就算明天遥远纷纷\n爱依然为彼此永恒\n\n【尾声】\n愿多年以后回头\n还能微笑着说——\n这是属于我们的温柔"
}
For variations, you can leave the sounds.text empty to let the AI create natural variations while preserving the essence of the original track.
Common Parameters
| Parameter | Type | Description |
|---|
action | string | Required. Operation type |
model | string | Required. Model to use |
sounds | array | Style prompts with strength and optional time range |
lyrics | string | Lyrics with section tags |
title | string | Song title (max 80 chars) |
lyricsStrength | number | Lyrics influence (0-1) |
weirdness | number | Creativity level (0-1) |
seed | number | Random seed for reproducibility |
makeInstrumental | boolean | Generate instrumental only |
Sound Item Properties
| Property | Type | Description |
|---|
text | string | Prompt describing the style |
strength | number | Strength (-1 to 1 for create, 0.2 to 1 for others) |
timeStart | number | Start time in seconds (create only) |
timeEnd | number | End time in seconds (create only) |
For detailed API specifications, explore the individual endpoint documentation.