Exploring the frontier of AI and human emotion — developing technologies that don't just process data, but understand it.
Our research in audio-based emotional intelligence analyzes vocal tones, stress levels, and context to recognize emotions in real time. Projects include Safety Zone distress detection and Clinical Context-Aware Emotion Recognition.
We developed a pipeline to guide LLMs in synthesizing insights from multiple documents without repetition or hallucination — ensuring context-aware accuracy and document harmony.
RootStock’s EI research fuses empathy with computation, exploring how AI can respond with emotional awareness — bridging the gap between logic and human connection.
An AI-driven system that evaluates musical performance and composition. Combining LLMs with grading algorithms, it assesses score sheets, identifies nuances, and offers constructive feedback.
Real-time emotion recognition from voice signals for safety and healthcare.
Document-aware synthesis ensuring consistency and reduced redundancy.
Assessing composition and performance with LLMs and grading algorithms.
We invite researchers, partners, and innovators to co-create the future of emotionally intelligent AI.
