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


Harmonic-percussive source separation

hpss(y, *[, kernel_size, power, mask, ...])

Decompose an audio time series into harmonic and percussive components.

harmonic(y, *[, kernel_size, power, mask, ...])

Extract harmonic elements from an audio time-series.

percussive(y, *[, kernel_size, power, mask, ...])

Extract percussive elements from an audio time-series.

Time and frequency

time_stretch(y, *, rate, **kwargs)

Time-stretch an audio series by a fixed rate.

pitch_shift(y, *, sr, n_steps[, ...])

Shift the pitch of a waveform by n_steps steps.


remix(y, intervals, *[, align_zeros])

Remix an audio signal by re-ordering time intervals.

trim(y, *[, top_db, ref, frame_length, ...])

Trim leading and trailing silence from an audio signal.

split(y, *[, top_db, ref, frame_length, ...])

Split an audio signal into non-silent intervals.

preemphasis(y, *[, coef, zi, return_zf])

Pre-emphasize an audio signal with a first-order differencing filter:

deemphasis(y, *[, coef, zi, return_zf])

De-emphasize an audio signal with the inverse operation of preemphasis():