Bookbot & University of Dar es Salaam

Learning by Doing: Research & Development Project

Overview

Bookbot, a digital reading tutor, integrates speech recognition technology with a comprehensive reading program tailored for school children to achieve enhanced literacy outcomes. We aim to introduce Bookbot to Kiswahili and make it accessible in Tanzania.

Aims

Developing Bookbot for Bahasa Indonesia, including a speech recognizer, text-to-speech, and a library of books, took a couple of years. Our goal is to accomplish the same in 3-6 months to explore how rapidly we can expand to other countries.

Why is this important?

Significant literacy challenges persist in sub-Saharan Africa and parts of Asia. Bookbot has demonstrated its effectiveness in improving literacy, with an efficacy study revealing a twofold increase in reading fluency.

Outputs

We are open-sourcing the Automatic Speech Recognition (ASR) and Text-to-Speech (TTS) to enable their utilisation in other Kiswahili EdTech software.

Project timeline

7 months

Current progress

We have just begun, and our initial ASR training shows a low error rate. Our TTS sounds natural, and we have successfully translated over 1,000 books for review. We are off to a promising start.

Meet the team

Adrian DeWitts is the Founder and Head of Technology at Bookbot.

Dr. Salome Maro is a Lecturer at the Department of Computer Science and Engineering at the College of Information and Communication Technologies (CoICT) at University of Dar es Salaam.

What is the biggest challenge you are currently facing?

The primary obstacles lie not in technology but in navigating bureaucracy. By collaborating with a university affiliated with the Ministry of Education, we hope to streamline this process.

What are you hoping to get from the community?

We seek to exchange ideas and share our knowledge. As a social enterprise, we feel good by doing good.

What measures are you taking to lower barriers to accessing AI in LMICs?

Once downloaded, Bookbot functions offline. The entire library, Text-to-Speech, and ASR are accessible without internet, allowing children to access extensive reading and narration resources. This has been invaluable for children in rural areas with limited internet connectivity.

How are you ensuring your AI tool is pedagogically sound?

We employ best practices in literacy learning and adhere to the country’s literacy curriculum. Our data indicates a twofold improvement in reading fluency, and our short-term efficacy study demonstrated progress compared to the control group.

Adapting Bookbot app from English to Swahili.

Pilot information
Audio: ASR
Large Language Models
ML: Classification
ML: Clustering
ML: Dimensionality Reduction
NLP: Generation

Based in Tanzania and Australia

U

Developing products for Tanzania

Contact – Bookbot

Adrian DeWitts

i

Founder and Head of Technology

Contact – Bookbot

Dr. Salome Maro

i

Founder and Head of Technology

Learning By Doing

We are providing small grants to support the development of AI products & components in LMICs. We know that innovation investment is high-risk. Our aim is that our community can benefit from the lessons learned in these pilots – what works and what doesn’t.

Learn more about our pilot projects here. We will be following each project and reporting on key learnings.

Sign Up

Join our mailing list to keep up to date with news and events.

Community          Knowledge          FAQ
          Privacy Policy

AI-for-Education.org was set up by Fab Inc. in partnership with Team4Tech. We are grateful to the Bill & Melinda Gates Foundation and the Jacobs Foundation for their support.

Powered by FabData.IO