Wonder Tech Reaches Milestone in Voice Screening for Depression
Wonder Technologies, a mental health solutions company based in Singapore, has reached a breakthrough in its acoustics-based artificial intelligence for depression screening. The AI model, which identifies depression by detecting nuanced vocal changes, has reached 83 percent in detection accuracy, surpassing the 80 percent threshold required for clinical depression detection.
By analyzing a 60-second voice sample, Wonder Tech's voice AI platform can detect and continuously monitor the risk of mental health conditions, categorized on varying levels of severity, to enable personalized treatment in a clinically reliable and scalable way.
Wonder Tech's AI model is acoustics-based rather than language-based. Since it does not rely on the content of speech and only requires a 60-second voice sample, the AI model can adapt to virtually all languages. It is hardware-independent, allowing implementation on consumer-grade smart devices, such as smartphones.
Depression can cause psychomotor function changes that impair cognitive and vocal cord functions. The effects are often exhibited in pitch, intensity, aspiration, jitter and other characteristics in voice. Wonder Tech's acoustics-based AI system can accurately assess for depression by detecting these subtle signs (objective biomarkers) using a large AI model with more than 300 million parameters.
To broaden the applications of its solution for greater impact, Wonder Tech is working on adapting its AI model to screen for various disorders, including anxiety, bipolar disorder and schizophrenia.
"Socioeconomic barriers continue to hinder access to treatment for the one billion people experiencing mental health disorders worldwide," saidWendy Wu, founder and CEO of Wonder Tech, in a statement. "In Singapore alone, anxiety and depression have been estimated to cost around $11.7 billion in lost productivity annually, the equivalent to 2.9 percent of the nation's GDP. Globally, depression and mental health disorders result in a staggering $1 trillion in lost productivity each year, underscoring an urgent need to improve their care systems."