You're looking at a specific version of this model. Jump to the model overview.
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 |
string
|
The audio file to separate
|
|
| model |
None
|
htdemucs
|
Model to use for separation (this deployment serves htdemucs)
|
| format |
string
|
wav
|
Output audio format, anything supported by FFmpeg
|
| isolate_stem |
None
|
none
|
Only creates a {stem} and no_{stem} stem/file
|
| shifts |
integer
|
1
Max: 20 |
Random time shifts to average; more is slower and slightly better. 0 disables them
|
| split_overlap |
number
|
0.25
Max: 0.99 |
Overlap between segments; higher values improve quality at segment boundaries
|
| seed |
integer
|
-1
|
Seed for the random shifts, for reproducible output; -1 for random
|
| clip_mode |
None
|
rescale
|
How to keep output within [-1, 1]; "none" leaves it unchanged
|
Output schema
The shape of the response you’ll get when you run this model with an API.
Schema
{'properties': {'bass': {'format': 'uri', 'title': 'Bass', 'type': 'string'},
'drums': {'format': 'uri', 'title': 'Drums', 'type': 'string'},
'no_bass': {'format': 'uri',
'title': 'No Bass',
'type': 'string'},
'no_drums': {'format': 'uri',
'title': 'No Drums',
'type': 'string'},
'no_other': {'format': 'uri',
'title': 'No Other',
'type': 'string'},
'no_vocals': {'format': 'uri',
'title': 'No Vocals',
'type': 'string'},
'other': {'format': 'uri', 'title': 'Other', 'type': 'string'},
'vocals': {'format': 'uri',
'title': 'Vocals',
'type': 'string'}},
'title': 'Output',
'type': 'object'}