analyticsinmotion/werpy

🐍📦 Ultra-fast Python package for calculating and analyzing the Word Error Rate (WER). Built for the scalable evaluation of speech and transcription accuracy.

61
/ 100
Established

This tool helps you quickly and accurately measure how well a spoken phrase has been converted into written text, or how similar two pieces of text are. By comparing a "reference" (the correct text) to a "hypothesis" (the transcribed or predicted text), it calculates the Word Error Rate (WER) and shows you exactly where mistakes like insertions, deletions, or substitutions occurred. It's designed for speech-to-text engineers, transcription quality analysts, and researchers evaluating text generation systems.

Used by 1 other package. Available on PyPI.

Use this if you need to objectively quantify the accuracy of speech recognition systems, transcription services, or text generation models by comparing their output against a correct version.

Not ideal if you only need a simple 'match/no match' comparison, or if your primary goal is spell-checking or grammatical correction rather than detailed error analysis of text transcription.

speech-to-text-evaluation transcription-quality-assurance natural-language-processing voice-ai-development text-accuracy-analysis
Maintenance 13 / 25
Adoption 7 / 25
Maturity 25 / 25
Community 16 / 25

How are scores calculated?

Stars

23

Forks

6

Language

Python

License

BSD-3-Clause

Last pushed

Mar 16, 2026

Commits (30d)

0

Dependencies

2

Reverse dependents

1

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/analyticsinmotion/werpy"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.