FeatureInput#

Defined in: voxatlas.features.feature_input

class voxatlas.features.feature_input.FeatureInput(audio, units, context)[source]#

Bases: object

Bundle the inputs passed to one extractor invocation.

Parameters:

audio (Audio | None) – Audio stream for the current conversation channel.
units (Units | None) – Hierarchical unit container for the current stream.
context (dict) – Shared execution context. The pipeline stores configuration and the feature store here.

Returns:

Dataclass instance used by extractors.

Return type:

FeatureInput

Notes

The context dictionary is the standard place for cross-feature state such as dependency outputs and runtime configuration.

Examples

Usage example:

feature_input = FeatureInput(audio=audio, units=units, context={"feature_store": store})
print(feature_input.context.keys())