MindVoice™ Voice-biomarker mental-health screening
Human Signals.
Health Intelligence.
MindVoice is consent-led voice-biomarker screening for workforce and student mental-health risk — distress, stress, burnout, fatigue and mental strain. Anonymised cohort signals for the organisation, a private self-care plan for every participant. Non-diagnostic. No individual health labels.
Non-diagnostic Consent-led Multimodal
Built in India. Designed for real-world diversity.
1,665validated participants
Ethics-approvedPhase 1, IEC protocol
Registered studyCTRI/2025/09/094592
Clinical Advisory Boardindependent review
NVIDIA Inceptionmember programme
MindVoice at a glance
MindVoice in one minute.
Brief, consent-led voice and multimodal check-ins that turn everyday interactions into private guidance for participants and anonymised cohort intelligence for organisations.
5 current signal areas1,665 validated participantsEthics-approved & clinically reviewed
Screening and signalling only. Not diagnosis, surveillance or performance management.
How MindVoice works
Moments become signals. Signals become decisions.
A brief interaction becomes a layered, non-invasive read across the five signals that matter now, with nine additional signals kept in research until the evidence is ready.
DistressStressBurnoutFatigueMental strain
Signal readFatigue pattern
Private supportSuggest recovery exercises
Organisation insightAnonymised cohort trend
Live voice demo
See the first signal layer form.
Use your microphone to reshape a local acoustic field, then see how screening becomes private guidance and cohort intelligence.
Voice-to-signal demoBrowser-only visualisation
Interactive signal fieldREADY
Try it in your browser
Sound becoming a signal.
This is the first layer only — a local visualisation of your voice. In a real pilot it becomes the anonymised cohort signal map and a private self-care plan for every participant.
Your browser will ask for microphone access. Audio is visualised locally and never leaves your device or reaches Mindei.
After screening
The participant gets value immediately.
Adoption improves when participants receive something useful after the check-in: a private self-care plan by email and access to practical psychological exercises such as box breathing.
01Screen
A brief consent-led check-in captures voice and other non-invasive signals.
02Understand
The participant receives a private summary written as guidance, not diagnosis.
03Practise
Self-care exercises such as box breathing, grounding and reflection are available after the screening.
04Support
Where needed, the pathway can point people toward EAP, counselling or agreed clinical support.
Exercises available after screening include box breathing, the physiological sigh, coherent breathing, grounding and reflection.
Use cases
Where earlier signal changes the decision.
Start with one cohort, one owner and one support pathway. Every pilot ends with a board-ready readout in six to ten weeks.
For Heads of People, Reward & Wellbeing
Your engagement survey reports last quarter's strain — after the attrition and absence have already been paid for.
With MindVoice. A voluntary 45–90 second voice check-in gives your wellbeing team an anonymised strain signal by team and function, weeks before a survey cycle would surface it. Every participant gets a private self-care plan, so people take part.
- Voluntary 45–90s check-ins — no individual health data shown to managers
- Strain signals by team, site and function
- Routing into your existing EAP and manager support
Also in the pilot readout
- Participation and completion, cohort by cohort
- A private self-care plan delivered to every participant
- A scale / refine / pause call with your executive sponsor
Also in scope
API & platform partners
Embed consented screening, participant guidance and anonymised cohort summaries into your own health, wellbeing or workforce product. Partner sandbox and governance support available.
Clinical & research
A governed screening-support layer that helps teams decide when someone may need the next level of care. Screening and signalling only — never diagnosis.
Enterprise pilot
Start with one cohort. Leave with a scale decision.
A focused 6-10 week pilot helps teams validate adoption, participant value and cohort-level signals before a wider rollout.
Choose one cohort
Start with a defined employee, student or care population and one clear business question.
Run consented check-ins
Collect brief voice-led interactions and give each participant private guidance after screening.
Review the cohort signal map
See anonymised patterns across the core wellbeing-risk signals, with privacy thresholds already built in.
Turn findings into action
Use the findings to tune wellbeing investment, referral pathways and leadership decisions.
Pilot readoutEvidence for a confident rollout decision.
The end-of-pilot readout shows whether MindVoice is ready to scale inside the organisation's population and support model.
Pilot delivers
A 6-10 week proof-of-value pilot gives the team the evidence needed to decide whether to scale.
- Participation and completion report
- Cohort signal dashboard
- Self-care plan emailed to each participant
- Exercise access such as box breathing
- Support-routing recommendations
- Executive pilot readout
Evidence
Designed to earn trust.
MindVoice has completed a clinician-supervised, ethics-approved Phase 1 programme in India, with a broader validation base of 1,665 participants and 40+ hours of voice-led signal data. Current product language stays focused on five validated signal areas; nine additional signals remain exploratory.
1,023Phase 1 participants
1,665validated participants
40h+voice-led signal data
CTRI2025/09/094592
Phase 1 was completed under an Institutional Ethics Committee-approved protocol.
Credibility
Evidence you can verify.
For health intelligence, trust is part of the product. MindVoice pairs clinical governance with privacy-first deployment and a clear AI ecosystem signal through the NVIDIA Inception Program.
Member of the NVIDIA Inception Program
A recognised startup programme for companies building with AI. Included here as technology ecosystem credibility, not as clinical endorsement.
Phase 1 was completed under an Institutional Ethics Committee-approved protocol.
Development is guided by clinical, behavioural science and AI expertise.
Built around consent, cohort thresholds and non-diagnostic screening boundaries.
From people who have taken a MindVoice screening
"MindVoice put into words what I couldn't figure out myself, making the experience feel simple and reassuring."
Veena JangirStudent, JIM
"The stress test provided valuable insights into stress levels, coping mechanisms, triggers and resilience."
Dr Akash DubeyFaculty, JIM
"The app, payment, test and results were smooth, and the AI counsel gave useful tips."
Dr Supriya KurlekarMD, Mai Mangeshkar Hospital, Pune
About Mindei Labs
Everyday human signals, translated into health intelligence.
Mindei Labs develops voice and multimodal biomarkers for health, wellbeing and human performance.
The technology analyses brief voice samples alongside other non-invasive signals to surface indicators associated with psychological distress, stress, burnout, fatigue and mental strain.
Our work combines AI, clinical research and behavioural science to turn everyday human signals into actionable insight for population-level screening, wellbeing support and human performance.
Meet the teamPrivacy by design
Built to support people, not surveil them.
The strongest sales story for health intelligence is trust. MindVoice is positioned around consent, cohort-level reporting and clear non-diagnostic boundaries from the first pilot conversation.
Employer views are designed around anonymised cohort intelligence.
MindVoice is not for hiring, firing, promotion, compensation or discipline.
Participants should know what is captured, why, and what support is available.
Signals should route to agreed wellbeing, EAP, counselling or clinical workflows.
Common questions
Clarity is part of the product.
Screening technology needs clear boundaries: what it does, where it helps and how it should be governed.
Give people support.
Give teams earlier signal.
Start with one cohort, five current signals and a clear screening-to-support pathway.
Best fit now500+ participants, consent-led deployment and an agreed support route.
Book a demo