Skip to main content

Python library for extracting chords from multiple sound file formats

Project description

chord-extractor

Python package

Python library for extracting chord sequences from sound files of multiple formats with the option of leveraging multiprocessing to source data from many files quickly. The extraction process wraps Chordino but is extensible to easily incorporate additional techniques.

Why?

  • Primarily intended for those analysing musical pieces and their harmonic progressions.
  • Chordino is a C++ Vamp Plugin for extracting chords but even with the helpful vamp Python wrapper, it is not trivial to set everything up. This project aims to help clarify the prerequisites and get the user up and running with extracting chords with as little fuss as possible.
  • Chord extraction of many files is time-consuming. This library gives the option of parallelization (on a particular machine) to cut the overall processing time considerably.
  • There are certain music files that are readily available but need converting prior to using the plugin (e.g. MIDI). This preprocessing is also included and can also be extended to convert other formats or other tasks that can take advantage of multiprocessing.

Installation

The package is hosted on PyPI, but prior to installing that there are a few prerequisite steps. The following instructions assume the latest versions of Ubuntu, and it is recommended to use a modern 64-bit Linux system. That said, equivalent steps should work if you are using another OS.

  • sudo apt-get install libsndfile1 - To read sound files.
  • (OPTIONAL) sudo apt-get install timidity - If wanting to extract chords from MIDIs (timidity converts midi to wav files).
  • (OPTIONAL) sudo apt-get install ffmpeg - If wanting to extract from mp3s
  • pip install numpy - numpy needs to be installed in your Python environment prior to installing chord-extractor. This is necessary as one of the package dependencies (vamp) requires it in its setup.py.

After that you are ready to run

pip install chord-extractor

NOTE: Included in the installation is a compiled library for Chordino. If you are using a Linux 64-bit OS, chord-extractor will default to using this binary. If you require a different version of the binary (i.e. you are using a different OS), please download the Vamp plugin pack installer for example if using another OS, please set the environment variable VAMP_PATH to point to the directory with the downloaded binary.

Usage

Extract chords from a single file:

from chord_extractor.extractors import Chordino

# Setup Chordino with one of several parameters that can be passed
chordino = Chordino(roll_on=1)  

# Optional, only if we need to extract from a file that isn't accepted by librosa
conversion_file_path = chordino.preprocess('/some_path/some_song.mid')

# Run extraction
chords = chordino.extract(conversion_file_path)
# => [  ChordChange(chord='N', timestamp=0.371519274), 
#       ChordChange(chord='C', timestamp=0.743038548), 
#       ChordChange(chord='Am7b5', timestamp=8.54494331),...]

To perform extraction of many files, even with various file types, we can pass a list of files in a single function. Here we can have 2 conversions running in parallel (preprocessing), and 2 extractions in parallel.

from chord_extractor.extractors import Chordino
from chord_extractor import clear_conversion_cache, LabelledChordSequence

files_to_extract_from = [
  '/path/file1.mid',
  '/path/file2.wav',
  '/path/file3.mp3',
  '/path/file4.ogg'
]

def save_to_db_cb(results: LabelledChordSequence):
    # Every time one of the files has had chords extracted, receive the chords here 
    # along with the name of the original file and then run some logic here, e.g. to 
    # save the latest data to DB

chordino = Chordino()

# Optionally clear cache of file conversions (e.g. wav files that have been converted from midi)
clear_conversion_cache()

# Run bulk extraction
res = chordino.extract_many(files_to_extract_from, callback=save_to_db_cb, num_extractors=2,
                            num_preprocessors=2, max_files_in_cache=10, stop_on_error=False)
# => LabelledChordSequence(
#	id='/tmp/extractor/d8b8ab2f719e8cf40e7ec01abd116d3a', 
#	sequence=[ChordChange(chord='N', timestamp=0.371519274), 
#	    ChordChange(chord='C', timestamp=0.743038548), 
#	    ChordChange(chord='Am7b5', timestamp=8.54494331),...])

If you want to implement your own extraction logic and/or add functionality to convert from another file format, whilst still taking advantage of the inbuilt multiprocessing logic, this can be done by extending the base class ChordExtractor

from chord_extractor import ChordExtractor, ChordChange
from typing import Optional, List
import os

class MyExtractor(ChordExtractor):
    def __init__(self, some_new_setting):
      self.some_new_setting = ####
    
    def preprocess(self, path: str) -> Optional[str]:
        conversion_path = super().preprocess(path)
        ext = os.path.splitext(path)[1]
        if ext in ['.newfmt']:
          # preprocess file at path, convert to .newfmt and have path to new temporary file
          conversion_path = ####
        return conversion_path
    
    def extract(self, filepath: str) -> List[ChordChange]:
        # Custom extraction logic using self.some_new_setting perhaps

For more documentation see here.

Contributing

Contributions, whether adding new functionality or raising an issue, are always welcome. You can see instructions on how to contribute in the CONTRIBUTING.md.

Project details


Download files

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

Source Distribution

chord-extractor-0.1.2.tar.gz (357.9 kB view hashes)

Uploaded Source

Built Distribution

chord_extractor-0.1.2-py3-none-any.whl (362.6 kB view hashes)

Uploaded Python 3

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