Check in on yourself in two minutes
Answer five short prompts on camera. The model reads your voice and facial motion together and shows you an emotion breakdown with four affective indicators — no questionnaire, no waiting room.
Your recordings stay in your account. Delete a check and the video goes with it.
Example result
An illustrative breakdown. A real check reports your own scores, the emotion distribution behind them and how each stage voted.
Why people use it
A second opinion on how you sound and look, taken as often as you want to take it.
Private by default
Recordings are tied to your account on a private disk. Nobody browses them, and deleting a check removes the file.
Two minutes, no forms
Record, submit and read the result. There is nothing to install and no questionnaire to fill in.
Your own history
Every check is kept in your list, so you can look back at how your results moved instead of judging a single day.
You see the reasoning
The emotion distribution, the per-stage predictions and the rule behind each indicator are all shown — not just a verdict.
How it works
Three steps, one short recording
-
1
Record
Answer five guided prompts on camera. The clip never leaves your account.
-
2
Analyze
Speech, facial motion and their fusion are scored by three trained networks.
-
3
Result
See the emotion distribution, indicator scores and how each modality voted.
What every account gets
The whole check, end to end — there is nothing held back behind an upgrade.
- Video checks from any device
- Full model breakdown on every result
- Your complete check history
- Arabic and English interface
Frequently asked questions
Who can see my recording?
Your check belongs to your account and is stored on a private disk — it is not published, shared or used to train the model. Platform administrators can see that a check exists for support and moderation purposes.
Can I delete a check?
Yes. Deleting a check from your list removes the stored result and the recording attached to it.
Is this a diagnosis?
No. It is a screening signal worded as a possibility. If a result worries you, take it to a qualified professional rather than acting on it alone.
Which language should I answer in?
Answer out loud in English. The speech network was trained on English, so another language weakens the audio stage and the fused result with it.
Try it on your own recording
Create an account and record two minutes — the result is ready in seconds.
Get StartedThis result is not a medical diagnosis. Please consult a professional.