librosax.feature.tempo¶

tempo(*, y: ndarray | None = None, sr: float = 22050, onset_envelope: ndarray | None = None, hop_length: int = 512, start_bpm: float = 120.0, std_bpm: float = 1.0, ac_size: float = 8.0, max_tempo: float = 320.0, aggregate: Callable | None = None, prior: Callable | None = None) → float | ndarray[source]¶

Estimate the tempo (beats per minute) from an audio signal.

This is a wrapper around librosa.feature.tempo.

Parameters¶

ynp.ndarray [shape=(…, n)] or None: Audio time series
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
start_bpmfloat > 0: Initial guess of the BPM
std_bpmfloat > 0: Standard deviation of tempo distribution
ac_sizefloat > 0: Length (in seconds) of the onset autocorrelation window
max_tempofloat > 0: Maximum tempo to consider (in BPM)
aggregatecallable or None: Aggregation function. If provided, used to combine tempo estimates.
priorcallable or None: Prior distribution over tempo