You're looking at a specific version of this model. Jump to the model overview.
deepsbhat1984 /voice-to-song:0e3d7ef8
Input schema
The fields you can use to run this model with an API. If you don’t give a value for a field its default value will be used.
| Field | Type | Default value | Description |
|---|---|---|---|
| audio_url |
string
|
Public URL of the vocal or song to cover (MP3/WAV/FLAC). "
"A cappella singing works - the melody is kept.
|
|
| prompt |
string
|
Target style for the cover: genre, instruments, mood, "
"vocal character. e.g. 'melodic trap, 808 bass, airy pads, "
"warm male vocal'. Describe the TARGET only, not the source.
|
|
| lyrics |
string
|
|
Optional new lyrics. Leave blank to keep/interpret the "
"source vocal. Keep syllable count near the source for timing.
|
| keep_melody |
number
|
0.75
Max: 1 |
How closely to follow YOUR melody (audio_cover_strength). "
"1.0 = closest to your source tune, lower = looser restyle. "
"0.7-0.85 keeps your melody while adding new production.
|
| steps |
integer
|
8
Min: 1 Max: 20 |
Diffusion steps. Turbo: 4-8.
|
| seed |
integer
|
-1
|
Random seed. -1 = random.
|
| audio_format |
None
|
mp3
|
None
|
| sample_rate |
None
|
44100
|
None
|
| bitrate |
None
|
256000
|
None
|
Output schema
The shape of the response you’ll get when you run this model with an API.
Schema
{'format': 'uri', 'title': 'Output', 'type': 'string'}