Spoken check-ins that help campus counsellors decide who to call first
TL;DR Quietcue lets students and employees answer a short wellbeing check-in by speaking in Hindi or English. A validated questionnaire scores it, and counsellors get a ranked call-back list. Colleges and employee-assistance providers pay per seat. Voice analysis stays a consented research track until it's proven in Indian languages.
India's national survey found 10.6% of adults living with a mental disorder at the time, and a treatment gap of 85% for common conditions like depression and anxiety[1]. Even where help exists, it queues. Tele-MANAS, the government helpline, has taken 47.75 lakh calls since launch[2].
A college counselling cell with two counsellors and a few thousand students has the same problem on a smaller scale.
Everyone who reaches out gets the same wait, and the counsellor only finds out who was most at risk after the call, which is far too late in the worst cases.
What it does, in regulatory terms: triage support for qualified counsellors, based on a validated questionnaire. No diagnosis, no treatment, no stand-alone voice screening.
A student or employee opens a link and answers PHQ-9 and GAD-7 items by speaking or tapping, in Hindi or English. Speech is transcribed, the answers are scored the standard way, and the counsellor's dashboard sorts people by severity and by how fast the score is changing. A high-risk answer shows the Tele-MANAS number straight away and alerts the counsellor.
The seed promised passive detection of mental illness from the voice alone, and I reframed it. A 2026 meta-analysis found that the most-studied voice measure, pitch, did not differ significantly between depressed and healthy groups, and warned that models may not carry across languages[3].
So Quietcue records voice features only with separate consent, for a validation study run with a psychiatry department. It shows no voice score until that study succeeds.
Users are students and employees. Payers are college counselling cells, universities and employee-assistance providers that already pay for counsellors and want them to see the right people first. There's no public Indian price list for this kind of tool. For reference, a consumer pays ₹3,749 a year for one wellbeing app's premium plan in India[4]. An institution should pay far less per head, so my assumption is ₹150–400 per seat a year.
CDSCO finalised its guidance on medical device software in July 2026, so founders now know where screening software crosses into device rules[5]. The DPDP Rules notified in November 2025 set out how consent for sensitive data has to work[6]. And speech recognition in Hindi and Indian English is now good enough to transcribe short spoken answers reliably.
There's no public figure for spending on counselling triage software in India. The demand-side signal is the gap itself - an 85% treatment gap[1] and 47.75 lakh helpline calls[2] show far more people asking than can be seen quickly.
Count your market bottom-up from colleges and employee-assistance providers in your first two cities.
AI wellbeing apps sell companions to consumers and employers and stick to wellness claims[4]. One US firm once known for voice-based depression screening, now markets an AI care-management calling agent to health plans[7], which suggests the stand-alone voice screen was a hard sell. Tele-MANAS is free but it's a helpline, not a tool for a campus counsellor.
The gap is a queue-ranking tool that sits under an institution's own counsellors.
| Difficulty | Moderate (48/100) |
| Starting capital | ₹60 L–1.2 Cr |
| Time to launch | 6–9 months |
The software is modest. The hard parts are clinical governance, getting consent right for sensitive health data, and convincing a counselling head that a ranked list won't miss someone. The voice research track needs a partner psychiatry department and ethics approval, and that takes months.
It may avoid device rules if it stays questionnaire-based triage support, but get that confirmed before launch.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
Don't trust AI to diagnose anyone or to decide someone is safe. It supports triage by qualified counsellors, and a low rank never replaces their judgement.
Want to work out the AI layer for Quietcue? Talk to Yash →
Here's what can go south -
I think it's worth building if you have a clinical psychologist co-founder and patience for ethics committees. If you want to sell an AI that hears depression then it is a poor fit. Prove one thing first - in a pilot, counsellors reach the highest-scoring tenth of students within 48 hours and say the list matched their own judgement.
Plan for roughly ₹60 L–1.2 Cr, covering clinical governance, consent design, the build and a partner research study.
Possibly. Screening or diagnostic software can be a medical device under CDSCO rules, so keep to questionnaire-based triage support and get a written opinion.
Yes, but validate the questionnaire in local languages and check local device, mental health and data protection law first.
The free Quietcue validation kit has the line-by-line capital breakdown, the effort scoring, the device-software and data-protection checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for counsellors and institutions, and 5 assumptions to test. Download the kit →
Thinking about building Quietcue or something close to it? The make-or-break work is earning a counselling head's trust that the ranked list is safe to act on, and there are no shortcuts there. 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.