LOS ANGELES, CA, UNITED STATES, January 19, 2026 /EINPresswire.com/ — Netverse introduced Raychel, a camera-free AI sleep companion designed to support sleep monitoring and touch-free interaction without collecting visual data.
As smart devices increasingly rely on cameras to interpret user behavior, concerns around privacy in personal spaces such as bedrooms have grown. Raychel adopts a non-visual sensing approach, using alternative technologies to detect user presence, movement, and interaction without image capture.
A Privacy-Oriented Approach to Sleep Companionship
According to Netverse, Raychel is designed to function as a sleep companion that combines task execution with personalized interaction. The system is built around an internal framework referred to as “Understand–Remember–Resonate,” which is intended to support contextual awareness, long-term personalization, and responsive interaction.
Non-Visual Sensing System Based on Radar and LiDAR
Raychel utilizes a multimodal sensing system that does not rely on cameras.
mmWave Radar for Contact-Free Sleep Monitoring
The device incorporates mmWave radar technology to enable continuous, contact-free tracking of body movement and respiration. Netverse states that this allows the system to determine whether a user is in bed, detect turning movements, monitor breathing rhythm, and estimate sleep-stage transitions, including REM sleep, without requiring wearable devices.
Because mmWave radar signals can penetrate bedding and clothing, sleep-related data can be collected without physical contact, reducing user discomfort.
TOF LiDAR for Touch-Free Interaction
Raychel also integrates a Time-of-Flight (TOF) LiDAR module to support high-frequency, real-time fine-motion detection, such as hand tracking. This capability enables gesture-based interaction, allowing users to perform actions such as snoozing alarms or adjusting lighting and audio playback without physical contact.
System Development and Technical Challenges
Netverse reports that the decision to exclude cameras required the development of proprietary signal processing and data fusion systems. Raw mmWave radar signals and LiDAR point-cloud data are processed through in-house algorithms to interpret user state and interaction intent.
By eliminating cameras at the hardware level, the system is designed to prevent image data collection, addressing privacy concerns associated with visual monitoring in private environments.
Dual-Module Intelligence Architecture
Raychel’s software system is built on a Mixture of Experts (MoE) architecture that separates conversational interaction from sleep data analysis.
One module focuses on personalized interaction and long-term preference learning, while a second module is dedicated to analyzing radar-based sleep data and presenting general wellness insights. Netverse emphasizes that Raychel is not a medical device, and that sleep-related feedback is provided for informational purposes only.
Future Development Toward On-Device Operation
Netverse also disclosed ongoing work to deploy language model capabilities directly on the device using the RWKV architecture. According to the company, future versions are expected to support reduced latency, local data processing, and operation without reliance on network connectivity.
Team Background
Raychel is developed by a team with more than ten years of experience in end-to-end product research, development, and management. The founding team includes specialists in sensing technologies and consumer hardware, with prior projects backed by investors such as ZhenFund, Sequoia Capital, and Lenovo Capital. Their experience spans service robotics, XR devices, and brain–computer interface hardware.
About Netverse
Netverse is a technology company focused on intelligent consumer devices, with an emphasis on non-visual sensing systems, embedded AI, and privacy-conscious design.
For more information, visit https://www.netverse.cc
CHEE
NETVERSE LIMITED
+1 908-733-1512
email us here
Legal Disclaimer:
EIN Presswire provides this news content “as is” without warranty of any kind. We do not accept any responsibility or liability
for the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in this
article. If you have any complaints or copyright issues related to this article, kindly contact the author above.
