rtvamp.compute_features¶

rtvamp.compute_features(timedata: ndarray, samplerate: float, plugin: str, *, blocksize: int | None = None, stepsize: int | None = None, parameter: dict[str, float] | None = None) → tuple[ndarray, list[ndarray]]¶

Compute features with plugin.

Parameters:

timedata – Time series data of arbitrary length
samplerate – Sampling rate in Hz
plugin – Plugin key/identifer as returned by e.g. list_plugins()
blocksize – Block size in samples. Default: preferred block size of plugin, otherwise 1024.
stepsize – Step size in samples. Default: preferred step size of plugin, otherwise = blocksize.
parameter – Dict with parameter identifiers and values. Use get_plugin_metadata() or Plugin.get_parameter_descriptors() to list available parameters and their constraints.

Returns: