NDP banners - website (6).png

Neural Decision Processors 

NDP101  - Always-On Sensor & Speech Recognition

 The Syntiant® NDP architecture is built from the ground up to run deep learning algorithms. Embedded with the Syntiant Core 1™, the NDP101 is a low-power, low-latency device that achieves 100x the efficiency levels of traditional CPUs and DSPs, all without requiring a connection to the cloud

Syntiant_NDP101_front_edited.png
Syntiant_NDP101_back.png

The NDP101 runs audio and sensor applications such as keyword spotting, wake word detection, speaker identification, sensor analytics and audio event and environment classification, while consuming less than 140μW of power. 

Ultra-Low-Power

Microwatt level active power consumption, while always-on and listening

Neural Processing

Neural network with half a million+ parameters providing large-scale parallelism for sensor applications

Keyword Speech

A keyword speech interface created from the simultaneous classification of dozens of keywords

System-on-chip

Programmable general purpose IO & SPI Master; booting from a serial flash with the ability to exist as the system's sole processor

Ease of Use

Multiple flexible methods to access neural processing engine with full customization of applications and post-processing

Key Features

  • System on chip in a 32 pin QFN package
     

  • Suitable for applications that require an MCU for control and a neural engine for decision making
     

  • Dual PDM microphone input or PCM-over-SPI input
     

  • Stereo/mono I2S interface multiplexed with PDM
     

  • Direct access to neural network over SPI for sensor applications
     

  • Frequency, time-domain & batch input models
     

  • 16-bit input holding tank with faster than real-time SPI extraction
     

  • Embedded Arm Cortex-M0 processor with 112KB SRAM
     

  • Eight GPIO pins with programmable direction and drive strength 
     

  • Support for external serial flash boot

  • Integrated clock multiplier and dividers support low-frequency clock source or external clocking
     

  • Optimized interrupt and SPI slave interface
     

  • Master SPI interface for sensor interfaces
     

  • Onboard firmware security and authentication
     

  • English speech service for keyword training
     

  • Software Development Kit (SDK) integrates in any software environment
     

  • Training Development Kit (TDK) to enable the use of standard frameworks such as TensorFlow for customer-programmed applications
     

  • Support for 64 output classifications
     

  • Active power consumption of <140µW while recognizing words

Contact us for resource requests & more information:

arrow&amp;v

Thank you for your message!  We'll be in touch.