IIT Bombay develops TARA app to boost reading fluency in schools

Representational Pic

In a fillip to the union school education ministries efforts towards addressing the learning crisis in India, IIT Bombay has developed an innovative mobile app, TARA (Teacher's Assistant for Reading Assessment), designed to assess and improve oral reading fluency (ORF) in students across the country.

The move comes at a time when the quality of education remains a major challenge, despite efforts to improve school attendance.

According to the Annual Status of Education Report (ASER) 2022, on an average more than half of Class 5 students in India struggle to read even simple Class 2-level texts, highlighting a major gap in foundational literacy.

Fire breaks out at Navi Mumbai building

1 hour agoWEBSTORY

The ASER reading test assesses whether a child can read letters, words, a simple paragraph of first grade level or a story of second grade level of difficulty.

The 2022 report also highlighted a concerning decline in reading levels among Class 3 students across both government and private schools. The percentage of students able to read text intended for second graders fell from 27.3% in 2018 to just 20.5% in 2022. This marks the lowest level since 2012, when only 21.5% of students demonstrated the same ability.

For fifth graders, the percentage of children able to read at least Class 2-level text decreased from 50.5% in 2018 to 42.8% in 2022.

Besides this NEP 2020 report further emphasized that over five crore students are yet to achieve foundational literacy and numeracy (FLN) skills. The COVID-19 pandemic only worsened the situation, with a staggering 90% of students losing specific language abilities, such as reading comprehension or picture description.

"As reading serves as a gateway to all other forms of learning, addressing this issue has become a priority for governments worldwide. In India, there is a concerted push to develop structured teaching methods and equip teachers with the tools to deliver effective education. However, one crucial area remains: monitoring learning outcomes. Regular assessments, particularly of oral reading fluency, are essential for measuring progress and guiding instructional strategies," read a statement issued by the IIT Bombay.

Traditionally, oral reading fluency assessments require one-on-one interaction with trained evaluators, where attributes like accuracy, speed, and smoothness are manually scored. However, this method is time-consuming and difficult to scale. To address this challenge, IIT Bombay's team of researchers, led by Professor Preeti Rao from the Department of Electrical Engineering, have developed TARA - a mobile app that automates the assessment of oral reading fluency through speech processing and machine learning technology.

TARA app works by analyzing an audio recording of a child reading an age-appropriate passage aloud. It automatically extracts critical rubrics such as words correct per minute (WCPM), and also assesses expression by measuring phrasing, intonation, and stress in speech. These factors are key indicators of the reader's understanding of the text. "The system is trained on expert-annotated recordings and currently supports English and Hindi, with its reliability matching that of human experts," said Rao.

The app addresses a long-standing need for real-time data on student performance, according to Dr. Shailaja Menon, reading pedagogy expert and Lead of the Center of Excellence in Early Language and Literacy at Tata Trusts.

TARA provides an end-to-end solution by enabling audio recordings and generating performance data for individual students, as well as for larger cohorts like classes, schools, and regions, all accessible via a user-friendly dashboard, Menon stated.

The project has garnered significant attention and support, receiving funding from the Tata Centre of Technology & Design and the Abdul Kalam Technology Innovation Fellowship. In a major milestone, the app has been adopted by the Kendriya Vidyalaya Sangathan (KVS) for assessing oral reading fluency in English and Hindi for Grades 3-8, covering over 7 lakh students across 1200 schools in India.

This initiative is now the largest oral reading fluency exercise ever conducted in the country.

The collaboration with KVS is particularly vital in light of the NIPUN Bharat National Mission, which aims to ensure that all students achieve FLN by the end of Grade 3.

KVS schools are expected to lead the way in this mission, serving as model schools for competency-based education and the adoption of learning outcome metrics.

For now the TARA team continues to work closely with KVS to develop effective remedial strategies for students, with the first round of assessments providing valuable data to refine teaching approaches.

Looking ahead, the success of this initiative will be gauged through regular test cycles and subsequent assessment phases, which will offer insights into how well the remediation strategies improve reading capabilities. "IIT Bombay is eager to expand partnerships and collaborations, with the ultimate goal of providing evidence-based solutions to transform the school education system in India," the institute's statement read.