Back close

ChalaChitra Vaachana: AI based Intelligent Audio Description Platform

Dept/Center/Lab: AmritaCREATE

Project Incharge: Prof. Dr. Prema Nedungadi

Center: AmritaCREATE

School: School of Computing

Collaborators: C-DAC Trivandrum

Funded by: MeitY, Government of India

Funded Amount: 18,20,17,000

Duration: 3 Years

 

ChalaChitra Vaachana: AI based Intelligent Audio Description Platform

Globally, at least 2.2 billion people experience near or distance vision impairment. As of 2022, approximately 4.95 million people in India are blind, which represents about 0.36% of the total population. Most video and audio content are not accessible to the blind, deaf and hard-of-hearing (DHH) people. To address this gap, the ChalaChitra Vaachana is an innovative audio description platform designed to improve the accessibility of visual content by automatically generating audio descriptions.

ChalaChitra Vaachana offers user-friendly web and mobile interfaces for generating intelligent audio descriptions. These descriptions are translated into Indian Sign Language (ISL) videos. The platform utilizes advanced AI models to accurately interpret scenes in videos. It supports a wide range of video formats and generates precise and detailed descriptions. The platform supports various output formats and includes features that enable content creators to review and edit descriptions. The project aligns with UN SDGs making educational content more accessible (SDG 4) and bridges existing accessibility gaps (SDG 10).

Key Highlights:

  • A fully functional platform that generates intelligent audio descriptors for videos, enhancing accessibility for blind and low-vision users.
  • A mobile version of the platform that provides accessibility and usability on mobile devices for viewing audio descriptions on the go.
  • Indian Sign Language (ISL) translation.
  • Advanced AI Models for audio descriptions.
  • Comprehensive dataset/corpus.
  • AI Models for ISL gesture translation.
  • The platform interface will be accessible to persons with disabilities (compliance with GIGW and accessibility guidelines).
  • Restriction on processing of obscene, anti-national, or objectionable content.

Related Projects

Development of Non-enzymatic Electrochemical Glucose Biosensors and Glucometer
Development of Non-enzymatic Electrochemical Glucose Biosensors and Glucometer
ConnectOne
ConnectOne
Bio-inspired Processor Design for Cognitive Functions via Detailed Computational Modeling of Cerebellar Granular Layer
Bio-inspired Processor Design for Cognitive Functions via Detailed Computational Modeling of Cerebellar Granular Layer
Feasibility of Plastic Brick Production for Rural Employment
Feasibility of Plastic Brick Production for Rural Employment
Modernization of Metallurgy Laboratory for Mechanical Testing, Metal Forming and Quantitative Metallurgical Microstructure Analysis
Modernization of Metallurgy Laboratory for Mechanical Testing, Metal Forming and Quantitative Metallurgical Microstructure Analysis
Admissions Apply Now