pychorus

Install the pychorus library using the standard Python package manager.

pip install pychorus

For deeper analysis, you can manually generate chromagrams and similarity matrices (Time-Time and Time-Lag) to visualize how different sections of the music relate to each other.

1. Use create_chroma to generate the chromagram and sample rate.
2. Initialize TimeTimeSimilarityMatrix and TimeLagSimilarityMatrix with the chroma and sample rate.
3. Call .display() on the matrix objects to visualize the results.

from pychorus import create_chroma
from pychorus.similarity_matrix import TimeTimeSimilarityMatrix, TimeLagSimilarityMatrix

# Generate chromagram
chroma, _, sr, _ = create_chroma("path/to/audio_file")

# Create matrices
time_time_similarity = TimeTimeSimilarityMatrix(chroma, sr)
time_lag_similarity = TimeLagSimilarityMatrix(chroma, sr)

# Visualize the results
time_time_similarity.display()
time_lag_similarity.display()

The simplest way to detect a chorus is to use the find_and_output_chorus function. This function identifies a chorus section in an audio file and saves it to a specified output file. It returns the starting time of the chorus in seconds.

from pychorus import find_and_output_chorus

# clip_length is the duration of the section to be extracted
chorus_start_sec = find_and_output_chorus("path/to/audio_file", "path/to/output_file", clip_length)

If you have cloned the repository, you can run the detection process directly from the terminal using main.py.

python main.py path/to/audio_file --output_file=path/to/output_file

The pychorus CLI allows you to identify the chorus of a music file and save it to a new audio file. It accepts a required input file path and provides optional flags to specify the output destination and the minimum duration required for a segment to be classified as a chorus.

Arguments:

• input_file: Path to the source audio file.

Options:

• --output_file: The path where the detected chorus will be saved. Defaults to chorus.wav.
• --min_clip_length: The minimum length in seconds for a segment to be considered a chorus. Defaults to 15 seconds.

# Basic usage (outputs to chorus.wav)
python main.py input.wav

# Specify custom output and minimum clip length
python main.py input.wav --output_file my_chorus.wav --min_clip_length 20