librosax.feature.zero_crossing_rate

zero_crossing_rate(y: Array, *, frame_length: int = 2048, hop_length: int = 512, center: bool = True, **kwargs) → Array

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

Parameters:

• y – Audio time series. Multichannel is supported.

• frame_length – Length of the frame over which to compute zero crossing rates

• hop_length – Number of samples to advance for each frame

• center – If True, frames are centered by padding the edges of y. Uses edge-value copies instead of zero-padding.

• **kwargs – Additional keyword arguments to pass to zero_crossings

Returns:

Zero crossing rate for each frame [shape=(…, 1, t)]

Return type:

jnp.ndarray