Is there an API available for this program?

[ibnYusrat]

Hi Guys, I have a strong feeling that I am asking a very silly question because the repo seems to be handling only the GUI part of the software and seems like there must be another repo that does the under-the-hood stuff, but if someone could please guide me in the right direction.

I want to know if its possible to run this program as a service where I could send it the inputs programmatically and receive the output files as a response.

Does such a thing exist for this. Any help in this regard would be highly appreciated.

[jocastrocUnal]

I like that to :)

[MohannadEhabBarakat]

I'm working on one here Now it supports 3 models only.

Planned features:

• [ ] Better docs
• [ ] Add rest of UVR_GUI models
• [ ] Auto testing
• [ ] Audio tools
• [ ] Pipelines to cascade models
• [ ] Ensembles to ensemble pipelines or models

I plan to release those features around mid of April. I'll keep this list updated

[terryops]

I'm working on one here Now it supports 3 models only.

Planned features:

• [ ] Better docs
• [ ] Add rest of UVR_GUI models
• [ ] Auto testing
• [ ] Audio tools
• [ ] Pipelines to cascade models
• [ ] Ensembles to ensemble pipelines or models

I plan to release those features around mid of April. I'll keep this list updated

Hello! Are you planning on completing this project? I can't see anything working in your repository.

[MohannadEhabBarakat]

I'm working on one here Now it supports 3 models only. Planned features:

• [ ] Better docs
• [ ] Add rest of UVR_GUI models
• [ ] Auto testing
• [ ] Audio tools
• [ ] Pipelines to cascade models
• [ ] Ensembles to ensemble pipelines or models

I plan to release those features around mid of April. I'll keep this list updated

Hello! Are you planning on completing this project? I can't see anything working in your repository.

You can use the models like this

import uvr
from uvr import models
from uvr.utils.get_models import download_all_models
import torch
import audiofile
import json

models_json = json.load(open("/content/ultimatevocalremover_api/src/models_dir/models.json", "r"))
download_all_models(models_json)
name = {name_of_your_audio}
device = "cuda"
    
demucs = models.Demucs(name="hdemucs_mmi", other_metadata={"segment":2, "split":True}, device=device, logger=None)

# Separating an audio file
res = demucs(name)
seperted_audio = res["separated"]
vocals = seperted_audio["vocals"]
base = seperted_audio["bass"]
drums = seperted_audio["drums"]
other = seperted_audio["other"]

I tested those on three models only but people from the committee tested multiple models and they worked fine (most of the time). I'm planning to test more models and add more functionalities (hopefully soon)

[3megabytesofhotram]

I've been using this https://pypi.org/project/audio-separator/