Wonder Tech is a Voice AI Platform for Mental Health. With its world-leading voice biomarker technology, the company empowers the full cycle of mental health solutions from voice mental health screening (SaMD) to monitoring and intervention. The urgent need for clinically validated and scalable mental health solutions has increased dramatically since the onslaught of the COVID-19 global pandemic. Supported by the latest FDA & NMPA policies in AI medical devices (SaMD) and digital therapeutics (DTx), the company co-creating a shared vision of the future of mental health that is empowered by its voice AI platform. Their secure, accurate, fast, and effective voice biomarker technology empowers a full cycle of mental health solutions from screening, monitoring, and personalized solutions for healthcare platforms, online hospitals, employee benefit programs, and insurance providers.

Voice production is a complex neuromuscular coordination process. Prior clinical research has shown that mental health disorders like depression affect the voice production process, for example, the voice from a depressed person was summarized as slow, monotonous, and disfluent with high jittering and shimmering. Such characteristics (or features/representations) in the voice, the so-called voice biomarker, can be used to assess or diagnose a condition/disease. Wonder Tech owns the most cutting-edge AI technology for depression assessment using voice biomarkers. To ensure high model accuracy, they follow the highest standard when collecting training data. Their multi-center research was designed and led by Peking University Sixth Hospital, one of the leading mental health institutes in China. Patients were diagnosed and recruited by psychiatrists from six different mental health hospitals across the country, following DSM-5 standards. Patients were given an H5 miniprogram for the voice sample collection, and the collection process was carefully designed, covering long vowels, number counting, rainbow passages, speech under cognitive load, open questions, etc. Our mental health dataset Oizys now contains more than 43000 audio sessions, collected from depression patients, anxiety patients, non-depression non-anxiety people, etc, and it’s by far the biggest audio dataset from DSM-5 diagnosed patients.

Leveraging the most advanced deep learning and transfer learning AI technology, Wonder Tech owns the most advanced AI model for depression assessment using voice biomarkers. Their AI model can give accurate assessment results based on 30-second voice recordings (16KHz, 16-bit). They first use self-supervised learning to learn latent voice feature representations from unlabeled voice data. These latent feature representations are then used as the input for another neural network which was trained on the Oizys dataset. Compared to AI models from traditional feature engineering (MFCCs etc.), our model achieves much higher performance (AUC 0.902) and provides better robustness in real-world scenarios.