Live Demo · Engine v1.0
Perceiving the unseen currents of human emotion.
Real-time multimodal monitoring — face, speech, and biosignals fused into one continuous baseline. Privacy-first, on-device.
Emotional State · Live
🙂
Peaceful
Confidence 94% · Negative affect 12%
"You seem calm and focused. Keep this balance with regular movement and hydration breaks."
94
score
Valence
0.82↑
Arousal
−0.14∼
Coherence
94%•
Trend
+8% vs yesterday
History
Calm62%
Neutral24%
Stressed14%
Recommend
5-min Breathing
Movement break
Hydrate
Advanced Analytics
Multimodal fusion timeline
1h24h7d30d
Face streamSpeech streamUpdated 0.2s ago
Methodology
Four modules. One continuous baseline.
Face stream
Frame-by-frame visual signal with on-device preprocessing.
Speech stream
Mel-spectrogram features over short rolling windows.
Late fusion
Weighted probability blending with single-modality fallback.
Trend analytics
Negative-affect score + threshold alerts with next steps.
Privacy first
Your raw video and audio never leave your device.
MAITRI processes camera and microphone frames locally and stores only derived emotion scores tied to your account.