Insta

Google India Launches Free And Open-Source Machine Learning Model 'MuRIL' For 16 Indian Languages

Swarajya Staff

Dec 18, 2020, 01:05 PM | Updated 01:05 PM IST


The Google office.
The Google office.

Tech gian Google has launched a machine learning tool for Indian languages to enable researchers, students, and startups to build local technologies in vernacular languages with a common framework, reports Economic Times.

The tool called Multilingual Representations for Indian Languages (MuRIL) has been developed by Google's India research unit, and it presently supports 16 languages and English. The tool aims to address concerns around Indian language understanding of computer systems, including all of its complexities like transliteration, spelling variations, mixed languages and other specific use cases which emerge in the Indian context.

Google claims that MuRIL offers the highest coverage for Indian languages among other such publically available models. Also, the tech major has made MuRIL free and open-source, making it available for download and use from its machine learning platform called TensorFlow.

Google has trained MuRIL tool using its existing language learning model called Bidirectional Encoder Representations from Transformers (BERT) which it uses to parse almost all the English queries on its search engine.

As per Partha Talukdar, a research scientist at Google India, MuRIL is a starting point of what the company believes can be the next big evolution for Indian language understanding.


Get Swarajya in your inbox.


Magazine


image
States