nemo_curator.stages.audio.metrics.get_wer

Module Contents

Classes

Name	Description
`GetPairwiseWerStage`	Count pairwise word-error-rate (WER) * 100% for each pair of text and pred_text.

Functions

Name	Description
`get_cer`	-
`get_charrate`	-
`get_wer`	-
`get_wordrate`	-

API

class nemo_curator.stages.audio.metrics.get_wer.GetPairwiseWerStage(
    name: str = 'GetPairwiseWerStage',
    text_key: str = 'text',
    pred_text_key: str = 'pred_text',
    wer_key: str = 'wer'
)

Dataclass

Bases: ProcessingStage[AudioTask, AudioTask]

Count pairwise word-error-rate (WER) * 100% for each pair of text and pred_text.

WER is measured between data[self.text_key] and data[self.pred_text_key].

Parameters:

text_key

strDefaults to 'text'

Key for the utterance transcript. Defaults to “text”.

pred_text_key

strDefaults to 'pred_text'

Key for the ASR predictions. Defaults to “pred_text”.

wer_key

strDefaults to 'wer'

Key to store the computed WER. Defaults to “wer”.

name

str = 'GetPairwiseWerStage'

pred_text_key

str = 'pred_text'

text_key

str = 'text'

wer_key

str = 'wer'

nemo_curator.stages.audio.metrics.get_wer.GetPairwiseWerStage.inputs() -> tuple[list[str], list[str]]

nemo_curator.stages.audio.metrics.get_wer.GetPairwiseWerStage.outputs() -> tuple[list[str], list[str]]

nemo_curator.stages.audio.metrics.get_wer.GetPairwiseWerStage.process(
    task: nemo_curator.tasks.AudioTask
) -> nemo_curator.tasks.AudioTask

nemo_curator.stages.audio.metrics.get_wer.get_cer(
    text: str,
    pred_text: str
) -> float

nemo_curator.stages.audio.metrics.get_wer.get_charrate(
    text: str,
    duration: float
) -> float

nemo_curator.stages.audio.metrics.get_wer.get_wer(
    text: str,
    pred_text: str
) -> float

nemo_curator.stages.audio.metrics.get_wer.get_wordrate(
    text: str,
    duration: float
) -> float