Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Feature] Create a Docker image for Apple Silicon #894

Open
vigneshsankariyer1234567890 opened this issue Mar 13, 2024 · 0 comments
Open
Labels
enhancement New feature or request feature New feature for Spleeter

Comments

@vigneshsankariyer1234567890

Description

At the moment, running the latest Docker image for Spleeter on Macs with Apple Silicon requires one of 2 things:

  1. Docker Desktop with at least v4.27.0, which runs Docker Engine v25.0.0,
  2. Specifying Docker engine to run on AMD64 platform

This feels really hacky, and essentially requires me to run the following command on my M1 Macbook Pro:

docker run --rm --platform linux/amd64 \
  -v "/path/to/input:/input" \
  -v "/path/to/output:/output" \
  -v "/path/to/model/volume:/model" \
  deezer/spleeter:3.8-5stems separate \
  -o /output \
  --codec mp3 \
  -p spleeter:2stems \
  /input/track.mp3

On top of that, when comparing execution speed against comparable source separation models such as demucs, performance is particularly affected even for separation on only 2 sources. Spleeter takes about 2 mins for a 30s clip to split into 2 sources, whereas Demucs takes only 1 min 20s to separate into 4 sources. It would be nice if there was an image compatible with ARM64. Of course, as an alternative option, we can also update the documentation on Docker to state the dependencies needed to run on Macs for local development.

@vigneshsankariyer1234567890 vigneshsankariyer1234567890 added enhancement New feature or request feature New feature for Spleeter labels Mar 13, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request feature New feature for Spleeter
Projects
None yet
Development

No branches or pull requests

1 participant