evidoria

← Back to browse

Good practice Imported

Read Along (Bolo) — Google's AI reading tutor for low-literacy children in India

India · Bengaluru · See the India profile · See the Bengaluru profile

Evidence: Quasi-experimental Top 10% 96/100 · Ask Evidence Copilot about this practice

Google's offline, speech-recognition reading tutor, launched in India as Bolo in 2019, gives children real-time voice feedback as they read aloud. A controlled pilot and an independent 2020 study both measured meaningful reading-fluency gains among beginner readers.

39 %
App users reaching the highest ASER reading level (pilot) (3 months)
28 %
Comparison-group children reaching the highest ASER reading level (pilot) (3 months)
64 %
Children with improved reading proficiency after 3 months (pilot) (3 months)
40 percentage points more
Additional users improving by one or more reading levels vs non-users (Sep 2020)
38% to 88%
Oral reading fluency among weakest beginner readers, before vs after ~100 minutes of use
7 million+
Cumulative student users (as of 2020 update)
Read Along (Bolo) — Google's AI reading tutor for low-literacy children in India

Details

Maturity
Established
Promoter
Google
Period
2019–present
Keywords
education, literacy, edtech, speech recognition, AI

Context

Google launched Bolo, later renamed Read Along, in India in March 2019: a free, offline Android app that uses on-device speech recognition and text-to-speech so children can practise reading aloud in Hindi or English. An animated companion, Diya, listens and gives spoken feedback, and adjusts story difficulty to the reader's level.

Results

In a pre-launch pilot, 64% of 920 children given the app improved their reading proficiency after three months, and 39% reached the highest ASER reading level, versus 28% of a 600-child comparison group. An independent September 2020 evaluation by Sattva Consulting, covering 3,500+ students across three languages and five Indian states, found 40% more Read Along users improved by at least one reading level than non-users, and oral reading fluency among the weakest beginner readers rose from 38% to 88% after roughly 100 minutes of use. By Google's 2020 update the app had reached over 7 million students who had cumulatively read more than 32 million stories, across 180+ countries and 10 languages.

Conclusions

Both the internal pilot and the Sattva evaluation measure only short-term reading-proficiency gains linked to app usage; no long-term, multi-year follow-up study was found in the public record.

Implementation
Indicative cost
Medium (€50k–€500k) — Google-funded product development, distributed free via the Android app store; no public budget figures reported.
Time to results
Long (> 3 years) — Launched March 2019 in India; by the 2020 update expanded to 180+ countries and 10 languages.
Staffing & skills
AI/ML engineers building on-device speech recognition, Content team producing levelled reading material in local languages

Conditions for success

  • Fully offline functionality for low-connectivity users
  • Story difficulty adjusted to the individual reader's level
  • Independent evaluation partner to verify gains (Sattva Consulting)

Where it fits

Governance type
private-sector product
Scale
national to global
Income level
low-to-middle-income focus

Commonly funded by

Own resources / municipal budget

Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Attachments

Similar practices you may find useful