Hindi reading catch-up for Classes 2–5 that flags who needs assessment
TL;DR Readcatch gives teachers in Hindi-medium schools a 20-minute daily small-group reading routine, built akshara by akshara, plus an app that listens to each child read aloud. Kids who stay stuck after a set number of weeks get flagged for formal assessment. Schools and CSR-funded programmes pay per child per year.
In rural India in 2024, only about 27% of Class 3 children could read a Class 2 text[1]. Somewhere inside that group are children with dyslexia. A meta-analysis puts specific learning disorders at 8% of Indian children, with dyslexia the most common type[2]. Can teachers tell the two groups apart? Rarely. Before training, Mumbai municipal teachers scored 24.7% on a test of learning-disability knowledge[3].
Diagnosis is slow and scarce. It needs a paediatrician, a psychiatrist and a psychologist, and the tests exist in only a few languages[4]. And most tools for dyslexic readers are built on English phonics, which simply doesn't fit Devanagari.
What it does, in regulatory terms: reading instruction and a referral flag only. Readcatch does not diagnose dyslexia or certify a disability, which under the RPwD framework needs a clinical team[4]. It does not sell therapy.
Teachers get a daily lesson script for groups of four to six children. It starts with the vowel signs and common consonants and then moves on to conjoint letters. The app shows the teacher what each child should practise next. A few times a week, each child reads a short passage into a shared phone, and speech recognition scores accuracy and speed. If a child does not improve after about 12 weeks of regular lessons then the app flags them and the school gets a referral letter for the parents.
The seed pitched AI reading adaptation for dyslexic students, and I'd reframe it. In Hindi-medium schools most struggling readers were never taught systematically, and very few dyslexic children are certified. Teaching all of them well and then watching who still struggles finds the dyslexic children and helps the rest. Early Kannada readers tune in to syllables before phonemes[5], which is why the lessons work at the akshara level.
The users are kids aged 7 to 10, but they don't pay. The payers are low-fee private schools in Hindi-belt towns and NGOs running CSR-funded reading programmes. A large literacy NGO's trial in Madhya Pradesh puts a benchmark on the table: ₹2,476 per child for each 0.1 SD of learning gain[6]. For comparison, one dyslexia charity's assistive reading app costs ₹600 a year[7] and one UK literacy app's English home plan starts at about ₹1,350 (£10.50) a month[8].
Section 16 of the RPwD Act already obliges schools to detect learning disabilities early and respond in the classroom[9]. The 2024 central guidelines start that screening at age eight or Class III[4]. Schools will soon have lists of flagged children and no Hindi programme to put them in, which is an awkward spot to be in. Reading scores also bounced back between 2022 and 2024[1], which shows that instruction can move quickly.
There is no public figure for spending on Hindi reading remediation. The firmer signals - about 73% of rural Class 3 children cannot yet read a Class 2 text[1], and NGOs already fund Hindi catch-up classes at known costs per child[6]. One Chennai dyslexia charity's Tamil programme shows how thin supply is, with four classrooms and 32 trained teachers[10].
A long-running dyslexia charity runs remediation in English and Tamil, plus an assistive reading app[7][10]. Large literacy NGOs run Hindi catch-up classes for all weak readers, with no dyslexia referral step[6]. Western adaptive literacy apps offer adaptive structured literacy, in English only[8][11]. I found no Hindi akshara-based remediation product with a referral step built in. That's the gap.
| Difficulty | Moderate (47/100) |
| Starting capital | ₹1.2–2.2 Cr |
| Time to launch | 8–12 months |
Hindi speech recognition for children reading aloud, in noisy classrooms and regional accents, is the hardest technical piece. You'll need to record and label your own audio. The other hard piece is making a busy teacher run 20 minutes a day with fidelity. A referral flag is only useful if the school then sends the child somewhere.
Readcatch needs no operating licence of its own, because it teaches reading and flags children for referral. The rules sit around it.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
AI should never decide that a child is dyslexic. It's a referral flag, and diagnosis stays with the clinical team.
Want to work out the AI layer for Readcatch? Talk to Yash →
I like this for a founder with foundational-literacy or NGO programme experience who can bring a speech-AI engineer on board. If you're hoping to sell a consumer dyslexia app to parents then it is a poor fit. Before you build much software, prove in 10 classrooms that children gain on a standard reading measure in one term, using paper lessons and a teacher's ear.
Plan for roughly ₹1.2–2.2 Cr, much of it on speech recognition built from your own labelled audio of children reading.
Not for reading instruction. You'd need RCI-registered staff to sell anything as therapy, and parental consent under the DPDP Act before you record children.
Yes, but the lessons are built for Devanagari, so another script means new lessons and new audio. Check your local disability and child-data laws too.
The free Readcatch validation kit has the line-by-line capital breakdown, the effort scoring, the disability, special-education and child-data checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for schools and NGOs, and 5 assumptions to test. Download the kit →
Thinking about building Readcatch or something close to it? The make-or-break bit is showing measurable reading gains in real Hindi classrooms before a funder will pay at scale, and that's the kind of thing I help founders with. If you need a hand, you know where to find me. Talk to Yash →
For illustration only. Names are fictional and ideas hypothetical; sourced facts are real but may be out of date. Nothing here is legal, financial or professional advice, and Yash Khatri is not personally liable for any use of it. Read the full disclaimer.