Beijing Zhongke Xinzhi has completed an angel funding round led by Shanghai Xinzhiyan. The company focuses on embodied intelligence emotion-sensing large models, fusing vision, voice, behavioral timing, and physiological signals to output clinical-grade empathy indices for next-generation human-machine interaction.
Funding Overview: Angel Round Closed, Focus on Affective AI
In August 2026, Beijing Zhongke Xinzhi Embodied Intelligent Technology Co., Ltd. announced the completion of its angel funding round, exclusively led by Shanghai Xinzhiyan Enterprise Management Partnership (Limited Partnership), with Wanchuang Investment Bank serving as the sole financial advisor.
Founded in February 2026, Zhongke Xinzhi has chosen a differentiated track within embodied intelligence — emotion perception. While the industry predominantly focuses on robotic motion control and visual recognition as "hard capabilities," Zhongke Xinzhi is dedicated to building the "emotional and social cognitive infrastructure" for next-generation embodied intelligence, enabling AI to transcend pure task execution and achieve understanding of human emotions, perception of states, and long-term human-machine empathetic interaction.
Technical Architecture: Four-Dimensional Cross-Modal Emotion Perception
Zhongke Xinzhi's core technical achievement is a cross-modal emotion-sensing large model that fuses four dimensions of data input:
- Visual signals: Identifying emotional states through facial micro-expressions and body language
- Voice signals: Extracting emotional information through analysis of intonation, speech rate, and timbre
- Behavioral timing: Capturing behavioral patterns and rhythm changes during interactions
- Physiological signals: Integrating heart rate, galvanic skin response, and other biometrics from wearable devices
The model employs Self-Attention cross-modal alignment mechanisms combined with clinical knowledge graphs to output clinical-grade "empathy indices" and multi-dimensional mental state vectors — meaning the system can deliver refined emotional metrics suitable for clinical assessment, far beyond simple "happy" or "sad" labels.
Application Scenarios: From Medical Rehabilitation to Smart Wearables
Zhongke Xinzhi's emotion perception technology targets multi-scenario deployment across:
- Medical healthcare: Post-operative patient emotion monitoring, early psychological disorder screening, elderly cognitive impairment assessment
- Education and training: Attention and emotion state tracking during learning, enabling personalized teaching interventions
- Elderly care companionship: Emotional companionship and mental health early warning for seniors living alone
- Smart wearables: Real-time emotion monitoring and health recommendations integrated into wearable devices
- Human-machine interaction: Emotional feedback and adaptive interaction for smart assistants and service robots
| Application Scenario | Core Value | Technical Support | Commercial Stage |
|---|---|---|---|
| Medical rehabilitation | Post-op emotion monitoring, early psychological screening | Cross-modal emotion recognition model | Product design and mass production underway |
| Education and training | Learning state tracking, personalized intervention | Behavioral timing + visual analysis | Standardized module R&D |
| Elderly care companionship | Emotional companionship, health alerts | Long-term emotional interaction modeling | Smart hardware commercialization |
| Smart wearables | Real-time emotion monitoring | Lightweight physiological signal fusion | Standardized sensing module deployment |
Capital Allocation: Three Pillars for Foundation Building
The angel round proceeds will be channeled into three key directions:
- Smart hardware commercialization: Translating emotion perception capabilities from algorithms to hardware products, accelerating mass production of smart devices
- Computing power expansion and model iteration: Scaling up training compute investments to continuously optimize the accuracy and generalization of the cross-modal emotion model
- Standardized sensing modules and compute cluster construction: Developing standardized emotion sensing modules to lower downstream integration barriers; building compute clusters to support model training and service deployment
Market Insight: The Industrialization Prospect of Affective AI
In 2026, the embodied intelligence sector has transitioned from the "how impressive is the body" demonstration phase to the "can it actually deliver" validation period. Primary market investment logic is shifting:
- From focusing on robotic body motion capabilities to questioning where physical interaction data comes from
- From competing on single-modal recognition accuracy to multi-modal fusion comprehensive intelligence
- From pursuing functional task completion to affective long-term interaction
Against this backdrop, emotion perception as the "soft capability" of embodied intelligence is moving from the periphery to the core. Accelerating global aging trends, increasing attention to mental health, and the deep integration of human-machine interaction into daily life — these macro trends collectively form the massive market foundation for affective AI. Although Zhongke Xinzhi has been established for only six months, its chosen track aligns closely with the rhythm of industrial evolution. As model capabilities continue to improve and hardware products gradually deploy, emotion perception is poised to become an indispensable middleware capability in the embodied intelligence value chain. For more frontier updates on embodied intelligence, visit EX1000.COM.
