nemo_curator.stages.audio.tagging

Audio tagging stages — processors for labeling unlabelled audio data.

Covers diarization, ASR alignment, and merge stages. Applicable across modalities (ASR, TTS, etc.).

Uses lazy imports to avoid loading heavy dependencies (NeMo, PyAnnote, etc.) until they are actually needed.

Subpackages

Name	Description
`__dir__`	List available attributes for autocomplete.
`__getattr__`	Lazy import handler - only imports modules when accessed.

nemo_curator.stages.audio.tagging.__dir__() -> list[str]

List available attributes for autocomplete.

nemo_curator.stages.audio.tagging.__getattr__(
    name: str
) -> typing.Any

Lazy import handler - only imports modules when accessed.

nemo_curator.stages.audio.tagging._LAZY_IMPORTS = {'ManifestReader': 'nemo_curator.stages.audio.common', 'ManifestReaderStage': 'n...

nemo_curator.stages.audio.tagging.__all__ = list(_LAZY_IMPORTS.keys())

nemo_curator.stages.audio.tagging._cache: dict[str, Any] = {}