A PyPI package for fast word/character error rate (WER/CER) calculation

• fast (cpp implementation)
• sentence-level and corpus-level WER/CER scores
• Unicode code point-based character error rates

pip install pybind11
pip install fastwer

import fastwer
hypo = ['This is an example .', 'This is another example .']
ref = ['This is the example :)', 'That is the example .']

# Corpus-Level WER: 40.0
fastwer.score(hypo, ref)
# Corpus-Level CER: 25.5814
fastwer.score(hypo, ref, char_level=True)

# Sentence-Level WER: 40.0
fastwer.score_sent(hypo[0], ref[0])
# Sentence-Level CER: 22.7273
fastwer.score_sent(hypo[0], ref[0], char_level=True)

Character-level scoring treats each Unicode code point as one character. Normalize text before scoring if canonically equivalent representations, such as composed and decomposed accented characters, should compare as equal.

Changhan Wang (wangchanghan@gmail.com)