The world’s leading programmable, ultra-low power neuromorphic chip for low-dimensional signal processing.
is a programmable, ultra-low power neuromorphic chip for low-dimensional signal processing. combines the analog front end(AFE) that can efficiently provide pre-processing functionality to input analog signals. is highly re-configurable and scalable, which supports feed-forward, recurrent and reservoir and other complex neural network structure. can be easily combined with various common sensors such as: MEMS microphone , thermal sensor, pressure sensor, vibration sensor, IMU, Gyro, PPG sensor etc.
- Audio detection
- Scene detection
- Keyword recognition
- Behavioral state detection
- Gait detection and analysis
- Breath detection
- ECG and EMG signal analysis
- Smart ear tag
- Biometric behavior recognition
- Border and coastal defense movement detection
- Glass break detection
- Industrial fault monitoring
Based on novel Spiking Recurrent Neural networks, reduce hardware resources by 10 times.
Utilizing the dynamic sparse signal and event-driven processing, 10x lower power consumption.
Novel chip architecture, 10x lower cost
No more caching, real time signal conversion and processing
Adaptive to various of sensors, excellent for temporal information processing.
Xylo-Audio Development Kit
It provides the capabilities for real-time audio detection. A high performance MEMS mic, power consumption measuring unit, external audio input port and a USB-C port are also provided by development kit.