Evaluate your speech-to-text system with similarity measures such as word error rate (WER)
-
Updated
Sep 3, 2026 - Python
#
word-error-rate
Here are 48 public repositories matching this topic...
An efficient OpenFST-based tool for calculating WER and aligning two transcript sequences.
-
Updated
May 12, 2026 - C++
한국어 STT 출력의 CER, WER, CRR, 키워드·개체명·코퍼스 평가를 제공하는 Python 패키지
aws amazon test speech-recognition openai korean speech-to-text evaluation-functions evaluation-metrics normalization cer transcribe evaluate speech-analysis wer word-error-rate text-evaluation text-digitisation character-error-rate computing-error-rates
-
Updated
Oct 2, 2026 - Python
Comprehensive voice cloning evaluation: seven automatic speech metrics and ready-to-use benchmark datasets. Score any model’s audio with Python or CLI.
benchmark text-to-speech tts speech-synthesis speaker-verification python-package evaluation-framework benchmark-datasets word-error-rate adversarial-audio voice-cloning speech-evaluation voice-clone zero-shot-tts audio-deepfake audio-protection speaker-similarity tts-evaluation speech-metrics voice-cloning-evaluation
-
Updated
Oct 20, 2026 - Python
Evaluate results from ASR/Speech-to-Text quickly
diff statistics evaluation comparison speech-recognition accuracy words speech-to-text stt difference asr mismatches wer word-error-rate transcriptions punctuations insertions
-
Updated
Dec 28, 2021 - JavaScript
🐍📦 Ultra-fast Python package for calculating and analyzing the Word Error Rate (WER). Built for the scalable evaluation of speech and transcription accuracy.
python nlp metrics levenshtein-distance automatic-speech-recognition speech-to-text stt asr python-package wer word-error-rate stt-benchmark asr-evaluation werpy
-
Updated
Oct 15, 2026 - Python
Calculates the word error rate of two strings, and the result is written into beautify HTML.
-
Updated
Mar 19, 2020 - Python
agent-oriented wer open metrics
-
Updated
Jul 27, 2026 - Python
🐍📦 Easy-to-use Python package for lightning-fast Word Error Rate (WER) analysis
metrics levenshtein-distance automatic-speech-recognition speech-to-text stt asr wer word-error-rate werx python-rust
-
Updated
Sep 8, 2026 - Python
Python tools to compare output transcript to reference
-
Updated
Jul 14, 2021 - Python
WhisperCPP (FP32) INT8, INT4, INT5, quantization effect on model latency and WER experiment
-
Updated
Oct 5, 2025 - Jupyter Notebook
A simple Python package to calculate word error rate (WER).
-
Updated
Feb 7, 2021 - Python
Open benchmark of 8 Bengali speech-to-text APIs (Sarvam, Soniox, Deepgram, Google Gemini & Chirp 2, ElevenLabs, OpenAI, Groq Whisper) on 1,001 real Bangla audio clips across 13 domains. Batch + streaming: CER, WER, latency and cost. By Riverborn.
benchmark gemini openai bengali bangla speech-to-text stt whisper asr word-error-rate bangla-nlp groq deepgram streaming-asr character-error-rate elevenlabs soniox bengali-asr sarvam-ai riverborn
-
Updated
Oct 14, 2026 - HTML
Developed a Marathi speech-to-text application using the Hugging Face whisper ASR models. Trained the model with a custom audio dataset and fine-tuned it for optimized performance. Deployed the model on the Hugging Face Model Hub, achieving a WER of 0.74 for the base model.
python ai voice pytorch artificial-intelligence voice-recognition speech-recognition openai automatic-speech-recognition speech-to-text whisper marathi asr voice-assistant wer word-error-rate huggingface marathi-language trasnformers gopiashokan
-
Updated
Jan 29, 2024 - Jupyter Notebook
Implementation of a couple of heuristics that estimate OCR quality without reliance on ground truth data, focusing on historical documents written in English.
-
Updated
Jun 2, 2023 - Python
Toolkit for using Whisper to transcribe YouTube videos. Includes Whisper transcription of YouTube videos, conversion of YouTube video into HuggingFace dataset (using audio and subtitles) and evaluation of Whisper transcription against YouTube subtitles
-
Updated
Oct 21, 2024 - Python
Word Error Rate computation using components from huggingface-evaluate and openai-whisper projects
-
Updated
Dec 30, 2024 - HTML
A word error rate util for golang
-
Updated
Nov 14, 2020 - Go
d-ser-t quantifies speech recognition accuracy of the MSFT speech service and/or user created MSFT custom speech service models.
-
Updated
Jan 4, 2023 - TypeScript
Add this topic to your repo
To associate your repository with the word-error-rate topic, visit your repo's landing page and select "manage topics."