doheejin/SB_loss_PA

This repository is the implementation of the paper, "Score-balanced Loss for Multi-aspect Pronunciation Assessment" (Interspeech 2023).

26
/ 100
Experimental

This project helps educators and language learning platform developers automatically assess speech quality. It takes recorded speech as input and provides detailed scores on pronunciation across various aspects. This tool is for researchers, speech pathologists, and language technology developers focused on improving automated pronunciation feedback.

No commits in the last 6 months.

Use this if you are developing or researching automated systems that need to give specific, nuanced feedback on spoken pronunciation.

Not ideal if you need a pre-packaged application for immediate end-user pronunciation practice without any development work.

pronunciation-assessment speech-technology language-learning computational-linguistics education-technology
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 6 / 25
Maturity 16 / 25
Community 4 / 25

How are scores calculated?

Stars

22

Forks

1

Language

Python

License

BSD-3-Clause

Last pushed

Apr 29, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/nlp/doheejin/SB_loss_PA"

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