kurianbenoy/malayalam_asr_benchmarking

A study to benchmark whisper based ASRs in Malayalam

37
/ 100
Emerging

This project helps developers and researchers evaluate the accuracy of speech-to-text models for the Malayalam language. By inputting specific Whisper or faster-Whisper based ASR models and a Malayalam speech dataset, it produces Word Error Rate (WER) and Character Error Rate (CER) metrics. This is useful for AI/ML engineers or computational linguists working on speech technology.

No commits in the last 6 months. Available on PyPI.

Use this if you are developing or selecting an Automatic Speech Recognition (ASR) model for Malayalam and need to compare different Whisper-based options rigorously.

Not ideal if you are a non-technical end-user simply looking to transcribe Malayalam speech without needing to evaluate model performance.

ASR-evaluation Malayalam-speech-recognition computational-linguistics AI-model-benchmarking speech-tech-development
Stale 6m
Maintenance 0 / 25
Adoption 5 / 25
Maturity 25 / 25
Community 7 / 25

How are scores calculated?

Stars

11

Forks

1

Language

Jupyter Notebook

License

MIT

Last pushed

Apr 15, 2024

Commits (30d)

0

Dependencies

10

Get this data via API

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

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