–

Interhuman AI raises €2M pre-seed round

–

Interhuman AI raises €2M pre-seed round

Social intelligence
for any AI application.

A single API that detects and interprets behavioral signals across video, audio, and text in real-time, enabling applications that adapt to how people actually communicate.

Inter-0

The first multimodal model for social intelligence.

Our first model Inter-0 detects agreement, confusion, engagement, hesitation, and 8 more observable behavioral signals - giving AI applications granular insights about the human on the other side of the screen.

How it works

Trained to understand context

Developer? Talk to us

Our model analyzes video, audio, and text together. Not as separate streams, but as integrated context.

It interprets the signals it detects, based on the context of the conversation. A pause means something different in negotiation training versus medical communication practice.

Hesitation

Built for applications where understanding behavior matters

Add behavioral feedback to training tools

Detect confidence, hesitation, and social signals during practice conversations. Provide feedback on delivery, presence, and reading the room - skills that determine outcomes. Measure communication capabilities traditional analysis can't capture.

Make learning experiences adapt to how users communicate

Analyze engagement, confusion, and communication effectiveness in real time. Create interactive practice environments that respond to behavioral signals. Transform passive content into adaptive experiences that drive measurable skill improvement.

Make agents respond to user behavior

Detect behavioral signals across voice, face, and body language during interaction. Enable your agent to adapt when users are confused, engaged, or hesitant. Respond to how people communicate, not just what they say.

Social intelligence infrastructure.

Inter-0 is the first multimodal model built specifically to detect behavioral signals in human communication. It's built for training scenarios where understanding how someone communicates matters as much as what they say.

Multimodal Perception

A single API that detects and interprets behavioral signals across voice, face, and body language.

Real-Time or Post-Processing

Respond during an interaction or analyze after. Same API, flexible deployment for your use case.

Built for Developers

REST API with clear documentation. Integrate in hours, get started with a few lines of code. Python and React SDKs available.

Grounded in Behavioral Science

Built on decades of research in human communication. Validated with psychologists.

Our Technology

⁂

Emotional Intelligence

Our AI simultaneously processes body language, facial expressions and voice patterns, detecting 7 distinct emotional states and emotional intensity.

✦

Real-time Multimodal Analysis

Real-time processing of behavioral, emotional, and contextual data, creating a rich understanding of human interaction that goes beyond words.

✲

Behavioral Analysis

Our system analyzes affect patterns, activity states, and contextual markers to detect subtle behavioral signals and interaction dynamics.

☉

Adaptive Response System

LLM-powered system adapts in real-time to emotional states and behavioral cues, delivering natural conversation flow with context-appropriate tone.

Our Research

Interpretability by design using computer vision for behavioral sensing in child and adolescent psychiatry

Using computer vision for behavioral sensing in child and adolescent psychiatry, this study assesses the accuracy of ML-derived behavioral codes from clinical interview videos, comparing them with human expert ratings to improve reliability and scalability in psychiatric diagnostics.

Read publication

Beyond Accuracy: Fairness, Scalability, and Uncertainty Considerations in Facial Emotion Recognition

The study examines the current state of FER models, highlighting issues of fairness, scalability, and robustness. The study proposes metrics and algorithms to assess and improve these aspects, emphasizing the importance of fair and reliable FER models in clinical applications and beyond.

Read publication

Scaling-up Behavioral Observation with Computational Behavior Recognition

The study proposes using open-source AI tools to automate behavioral coding in parent-child interactions and therapy sessions. This method enhances scalability, consistency, and depth of analysis, addressing traditional human coding limitations. The study discusses privacy, bias, and validation methods, highlighting the potential for these tools in psychological research and clinical practice.

Read publication