Article complete

IISc to Open Source Project Vaani Under Bhashini with 16,000 Hours of Speech Data

The ambitious project aims to curate datasets of 150,000 hours of natural speech and text from approximately one million people across 773 districts in India.

A map of India filled with various languages, highlighting linguistic diversity across the regions.

Key takeaways

  • IISc opens 16,000 hours of speech data for Project Vaani under the Bhashini initiative.
  • Project Vaani aims to collect 150,000 hours of speech data from one million participants across India.
  • Phase two will gather 200 hours of speech data from 160 districts, expanding language diversity.
  • Collaboration with Google enhances data quality and accessibility for AI development in India.
  • Focus on inclusive technology aims to assist people with impairments and improve education.

Takeaways drafted with AI from the article text.

The Indian Institute of Science (IISc) AI and Robotics Technology Park (ARTPARK) is set to open-source 16,000 hours of spontaneous speech data from 80 districts as part of Project Vaani, under the Ministry of Electronics and Information Technology's flagship AI initiative, BHASHINI. 

This effort is in collaboration with the Google.

The ambitious project aims to curate datasets of 150,000 hours of natural speech and text from approximately one million people across 773 districts in India. The first phase of the project, launched at the end of 2022, is nearing completion.

In its second phase, Project Vaani will target 160 districts, collecting 200 hours of speech data from about 1,000 people per district. So far, voice data in 58 different language variants or dialects has been gathered from 80 districts and will soon be made publicly available.

Prasanta Ghosh, Assistant Professor in the Department of Electrical Engineering at IISc, leading Project Vaani, mentioned, “Our journey in the Indic language data sets started in 2020 with RESPIN (recognising speech in Indian languages).”

Apart from healthcare, Ghosh is also focused on using AI for education in India on his project called English Gyani. Funded by the Ministry of Science and Technology, it is also called the IMPRINT project. “We’re building tools to help young jobseekers learn spoken English and comprehension skills right from their homes,” said Ghosh.

“Bhashini, for the first time, helped create the digital data for low resource languages in its effort to build AI models for low resource languages,” said Amitabh Nag, Chief Executive of Bhashini. 

The primary goal is to use this dataset as training data for speech-to-text AI models, particularly benefiting conversational AI platforms and chatbots requiring diverse voice datasets. Organisations can leverage this data to develop speech models for various Indian languages and dialects, aligning with their business objectives.

A Google spokesperson highlighted the project's significance, stating, "Project Vaani was born out of the deep need to make high-quality, diverse & anonymised datasets available for people to build technology solutions that reflect the way local languages are spoken. We are grateful for the tireless work of our partners to realise this commitment, & are very pleased with the ongoing momentum. We look forward to sharing more updates soon."

Published

Spotted an error? Tell our editors. Read our editorial standards and corrections.

More in this story

Open Source AI · 5 stories
  1. MIT Media Lab Open Sources AI-generated Characters Library
  2. Allen Institute for AI Releases OLMo, Open Source 7B and 1B Models
  3. You are here: IISc to Open Source Project Vaani Under Bhashini with 16,000 Hours of Speech Data
  4. NVIDIA Launches Ising Family of Open-Source AI Models for Quantum Computing
  5. Ministry of Ayush Partners with BHASHINI to Expand Multilingual AI Healthcare Services
About the author

Journalist

Mohit writes about AI in simple, explainable, and often funny words. He's especially passionate about chatting with those building AI for Bharat, with the occasional detour into AGI.

All stories
Belamy · Daily

The day’s AI news, before your first meeting.