Skip to main content

Automatically find loop points of any song and play endlessly or export into intro/loop/outro sections or loop points.

Project description

PyMusicLooper

A script for repeating music seamlessly and endlessly, by automatically finding the best loop points.

Features:

  • Find loop points within any music file (if they exist).
  • Supports a large set of different audio formats, and all the popular ones (MP3, OGG, M4A, FLAC, WAV, etc).
  • Play the music file endlessly and seamlessly with the best discovered loop.
  • Export to intro/loop/outro sections for editing or seamless playback within any music player that supports gapless playback.
  • Export loop points in samples (e.g. for use in creating custom themes with seamlessly looping audio).

Installation

Requires Python >=3.6 to run. This program depends on NumPy (for arrays and mathematical operations) and Librosa (for audio analysis and beat extraction). If you don't have these dependencies installed, they'll be automatically downloaded:

pip install git+https://github.com/arkrow/PyMusicLooper.git

To play music endlessly through the terminal, the external library mpg123 is required. Available through the following link: (https://www.mpg123.de/download.shtml).

Usage

usage: python -m pymusiclooper [-h] [-p] [-e] [-j] [-b] [-r] [-o OUTPUT_DIR]
                               [-m MIN_DURATION_MULTIPLIER]
                               path

Automatically find loop points in music files and play/export them.

positional arguments:
  path                  path to music file.

optional arguments:
  -h, --help            show this help message and exit
  -p, --play            play the song on repeat with the best discovered loop
                        point (default).
  -e, --export          export the song into intro, loop and outro files (WAV
                        format).
  -j, --json            export the loop points (in samples) to a JSON file in
                        the song's directory.
  -b, --batch           batch process all the files within the given path
                        (usage with export args [-e] or [-j] only).
  -r, --recursive       process directories and their contents recursively
                        (usage with [-b/--batch] only).
  -o OUTPUT_DIR, --output-dir OUTPUT_DIR
                        specify the output directory (defaults to the track's
                        directory).
  -m MIN_DURATION_MULTIPLIER, --min-duration-multiplier MIN_DURATION_MULTIPLIER
                        specify minimum loop duration as a multiplier of song
                        duration (default: 0.35)

PyMusicLooper will find the best loop point it can detect, and will then, depending on your arguments:

(a) play the song on repeat using the best discovered loop point (default, requires mpg123);

(b) export intro/loop/outro sections of the song (currently outputs as WAV-only, although you may convert with ffmpeg or Audacity);

(c) export the loop points (in samples) to a JSON text file, which you can use for audio loops in custom theme creation, etc.

Example Usage

Note: If on Windows, you can Shift+Right-Click in an empty spot in the song's folder and choose command-line/powershell from the context menu. Otherwise, cd/dir to the folder.

Play the song on repeat with the best discovered loop point.

python -m pymusiclooper "Song I Could Listen To Forever.mp3"

Export the song into intro, loop and outro files, as well as the loop points used (outputs in the same directory/folder as the track).

python -m pymusiclooper -ej "some music track.ogg"

Export the loop points of all the songs in the current directory.

python -m pymusiclooper -bj .

The I WANT IT ALL option. Export intro/loop/outro sections and loop points of all the songs in the current directory and its subdirectories, to a folder called "Music Loops".

python -m pymusiclooper -brej . -o "Music Loops"

If the loop is very long (or very short), you may specify a different minimum duration for the algorithm to use, which is 0.35 (35%) by default. If the most of the track is the loop section, specifying a higher multiplier will also speed the algorithm up. Here -m 0.85 means that, excluding silence, the loop section is at least 85% of the music track.

python -m pymusiclooper "super long track.flac" -m 0.85

Building from source

Requried python packages: pip and setuptools.

Clone the git repository to a directory of your choice and cd to inside the repo.

Run:

python setup.py build

Followed by:

python setup.py install

Contribution

If there is a song that you think PyMusicLooper should be able to loop but doesn't, please feel free to open an issue with a link to that song and mention the approximate timestamp at which it loops. Forks and pull requests are of course welcome.

Acknowledgement

This project started out as a fork of Nolan Nicholson's project Looper. Although at this point only a few lines of code remain from that project due to adopting a completely different approach and implementation, without their contributions this project would not have been possible.

Version History

  • v1.2.0 Removed unreliable cache implementation
  • v1.1.0 Added support for batch processing
  • v1.0.0 Initial Release

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distributions

No source distribution files available for this release.See tutorial on generating distribution archives.

Built Distribution

pymusiclooper-1.2.0-py3.7.egg (18.8 kB view details)

Uploaded Source

File details

Details for the file pymusiclooper-1.2.0-py3.7.egg.

File metadata

  • Download URL: pymusiclooper-1.2.0-py3.7.egg
  • Upload date:
  • Size: 18.8 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.3.0 pkginfo/1.7.0 requests/2.25.1 setuptools/52.0.0.post20210125 requests-toolbelt/0.9.1 tqdm/4.55.1 CPython/3.8.5

File hashes

Hashes for pymusiclooper-1.2.0-py3.7.egg
Algorithm Hash digest
SHA256 c9c304557e0efe5dbb151e6d69a51b180feb2cc31a4afae877f74bd8690c3fa5
MD5 71494af2353dbb7e724c85f6f3986d46
BLAKE2b-256 0e305ef5d4ad441ab70a0edaf3cbddf5f28e980ec24868b84a94ec53cc28c208

See more details on using hashes here.

Provenance

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page