Caution

You're reading the documentation for a development version. For the latest released version, please have a look at 0.10.1.

Feature extraction

Spectral features

chroma_stft(*[, y, sr, S, norm, n_fft, ...])

Compute a chromagram from a waveform or power spectrogram.

chroma_cqt(*[, y, sr, C, hop_length, fmin, ...])

Constant-Q chromagram

chroma_cens(*[, y, sr, C, hop_length, fmin, ...])

Compute the chroma variant "Chroma Energy Normalized" (CENS)

chroma_vqt(*[, y, sr, V, hop_length, fmin, ...])

Variable-Q chromagram

melspectrogram(*[, y, sr, S, n_fft, ...])

Compute a mel-scaled spectrogram.

mfcc(*[, y, sr, S, n_mfcc, dct_type, norm, ...])

Mel-frequency cepstral coefficients (MFCCs)

rms(*[, y, S, frame_length, hop_length, ...])

Compute root-mean-square (RMS) value for each frame, either from the audio samples y or from a spectrogram S.

spectral_centroid(*[, y, sr, S, n_fft, ...])

Compute the spectral centroid.

spectral_bandwidth(*[, y, sr, S, n_fft, ...])

Compute p'th-order spectral bandwidth.

spectral_contrast(*[, y, sr, S, n_fft, ...])

Compute spectral contrast

spectral_flatness(*[, y, S, n_fft, ...])

Compute spectral flatness

spectral_rolloff(*[, y, sr, S, n_fft, ...])

Compute roll-off frequency.

poly_features(*[, y, sr, S, n_fft, ...])

Get coefficients of fitting an nth-order polynomial to the columns of a spectrogram.

tonnetz(*[, y, sr, chroma])

Compute the tonal centroid features (tonnetz)

zero_crossing_rate(y, *[, frame_length, ...])

Compute the zero-crossing rate of an audio time series.

Rhythm features

tempo(*[, y, sr, onset_envelope, tg, ...])

Estimate the tempo (beats per minute)

tempogram(*[, y, sr, onset_envelope, ...])

Compute the tempogram: local autocorrelation of the onset strength envelope.

fourier_tempogram(*[, y, sr, ...])

Compute the Fourier tempogram: the short-time Fourier transform of the onset strength envelope.

tempogram_ratio(*[, y, sr, onset_envelope, ...])

Tempogram ratio features, also known as spectral rhythm patterns.

Feature manipulation

delta(data, *[, width, order, axis, mode])

Compute delta features: local estimate of the derivative of the input data along the selected axis.

stack_memory(data, *[, n_steps, delay])

Short-term history embedding: vertically concatenate a data vector or matrix with delayed copies of itself.

Feature inversion

inverse.mel_to_stft(M, *[, sr, n_fft, power])

Approximate STFT magnitude from a Mel power spectrogram.

inverse.mel_to_audio(M, *[, sr, n_fft, ...])

Invert a mel power spectrogram to audio using Griffin-Lim.

inverse.mfcc_to_mel(mfcc, *[, n_mels, ...])

Invert Mel-frequency cepstral coefficients to approximate a Mel power spectrogram.

inverse.mfcc_to_audio(mfcc, *[, n_mels, ...])

Convert Mel-frequency cepstral coefficients to a time-domain audio signal