S P I R E    L A B

About

There are thousands of languages and dialects in India that are actively spoken. Many regional Indian languages have minimal speech data on the internet, making it difficult to develop accurate AI and voice technologies for them. Government of India through Ministry of Electronics and Information Technology (MeiTy) started an initiative called National Language Translation Mission (NLTM) with a goal of making technology accessible in all Indian languages. SPIRE Lab's project, ASR for Extremely Low-Resource Indic Languages supports NLTM's objective by gathering data for three low resource languages languages

Objectives:

Our project focuses on building Automatic Speech Recognition (ASR) systems for three low-resource Indian languages.

• Obtain speech data and transcription for Maithili, Santali and Konkani
• Develop ASR models for the three languages
• Evaluate the ASR models

>>top<<