FICHA · AUR

python-jiwer

Evaluate your speech-to-text system with similarity measures such as word error rate (WER).

  • Speech recognition metric library
  • LIBRARY
  • AUDIO
  • AI
  • Dependency only
official+codex · reviewed · Jun 3, 2026 description in en

Description

Speech-to-text outputs can be evaluated with word error rate and related similarity measures. Researchers and accessibility developers use jiwer to compare transcripts against references. Reference text and transcripts may contain private speech, so evaluation data needs care.

Permissions

Permissions not analysed for this source yet.