librosax.feature.tempogram¶

tempogram(*, y: ndarray | None = None, sr: float = 22050, onset_envelope: ndarray | None = None, hop_length: int = 512, win_length: int = 384, center: bool = True, window: str = 'hann', norm: float | Callable | None = inf) → ndarray[source]¶

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

This is a wrapper around librosa.feature.tempogram.

Parameters¶

ynp.ndarray [shape=(…, n)] or None: Audio time series. Required if onset_envelope is not provided.
srfloat > 0: Sampling rate of the audio time series
onset_envelopenp.ndarray [shape=(n,)] or None: Optional pre-computed onset strength envelope
hop_lengthint > 0: Number of audio samples between successive onset measurements
win_lengthint > 0: Length of the onset autocorrelation window (in frames/onset measurements)
centerbool: If True, onset autocorrelation windows are centered. If False, windows are left-aligned.
windowstr, tuple, number, callable, or list-like: A window specification as in get_window
norm{np.inf, float > 0, None, callable}: Normalization mode. Set to None to disable normalization.