medkit.audio.preprocessing#

APIs#

For accessing these APIs, you may use import like this:

from medkit.audio.preprocessing import <api_to_import>

Classes:

`Downmixer`(output_label[, prevent_clipping, uid])	Downmixing operation converting multichannel audio signals to mono.
`PowerNormalizer`(output_label[, ...])	Normalization operation setting the RMS power of each audio signal to a target value.

class Downmixer(output_label, prevent_clipping=True, uid=None)[source]#

Downmixing operation converting multichannel audio signals to mono.

Parameters

output_label (str) – Label of output downmixed segments.
prevent_clipping (bool) – If True, normalize downmixed signals by number of channels to prevent clipping.
uid (str) – Identifier of the downmixer.

Methods:

`run`(segments)	Return a downmixed segment for each segment in segments.
`set_prov_tracer`(prov_tracer)	Enable provenance tracing.

Attributes:

description

Contains all the operation init parameters.

run(segments)[source]#

Return a downmixed segment for each segment in segments.

Parameters: segments (List[Segment]) – Audio segments to downmix.
Return type: List[Segment]
Returns: List[~medkit.core.audio.Segment] – Downmixed segments, one per segment in segments.

Contains all the operation init parameters.

set_prov_tracer(prov_tracer)#

Enable provenance tracing.

Parameters: prov_tracer (ProvTracer) – The provenance tracer used to trace the provenance.

class PowerNormalizer(output_label, target_value=1.0, channel_wise=False, uid=None)[source]#

Normalization operation setting the RMS power of each audio signal to a target value.

Parameters

output_label (str) – Label of output normalized segments.
target_value (float) – Value to set the RMS power of each segment to.
channel_wise (bool) – If True, the normalization is performed per-channel, thus modifying the balance of multichannel signals.
uid (str) – Identifier of the normalizer.

Methods:

`run`(segments)	Return a normalized segment for each segment in segments.
`set_prov_tracer`(prov_tracer)	Enable provenance tracing.

Attributes:

description

Contains all the operation init parameters.

run(segments)[source]#

Return a normalized segment for each segment in segments.

Parameters: segments (List[Segment]) – Audio segments to normalize.
Return type: List[Segment]
Returns: List[~medkit.core.audio.Segment] – Power-normalized segments, one per segment in segments.

Contains all the operation init parameters.

set_prov_tracer(prov_tracer)#

Enable provenance tracing.

Parameters: prov_tracer (ProvTracer) – The provenance tracer used to trace the provenance.

`medkit.audio.preprocessing.downmixer`
`medkit.audio.preprocessing.power_normalizer`
`medkit.audio.preprocessing.resampler`	This module needs extra-dependencies not installed as core dependencies of medkit.