Highly Efficient AI Semiconductors to Enable Products with Alexa Built-In
Irvine, Calif., August 5, 2019 - Syntiant Corp., an artificial intelligence semiconductor company backed by some of the world’s leading tech companies, today announced its Syntiant™ NDP100™ and Syntiant NDP101™ microwatt-power Neural Decision Processors™ (NDPs) are now available for use with the Amazon Alexa Voice Service (AVS).
As a low-power processor for close-talk voice-enabled devices, Syntiant’s neural network technology allows OEMs to offer a built-in Alexa experience to even the most power sensitive battery-powered devices. From ear buds to laptops, consumers can enjoy hand-free interactions with Alexa, including new features and access to tens thousands of Alexa skills.
“Our NDP10x series of neural decision processors are a new type of semiconductor for running deep learning algorithms,” said Kurt Busch, CEO of Syntiant. “These chips are purpose-built for keyword spotting such as wake words like Alexa, and now our processors can be used for quickly developing voice applications in battery-powered devices.”
Custom built to run neural workloads, the NDP100 and NDP101 consume less than 150 microwatts while continuously listening for the corresponding wake word. Syntiant’s processors enable devices to be approximately 200x more efficient, providing 20x the throughput of the typical low-power MCU solution.
“Syntiant’s neural decision processor technology takes a unique approach to wake word detection on voice-enabled devices,” said Sanjay Voleti, senior manager of device enablement with the Alexa Voice Service. “We recently conducted acoustic performance tests of their latest solutions, and were impressed by their ability to support compute-on-edge and wake word detection on low-power devices. We’re excited to see developers begin using this technology in their devices and deliver new Alexa experiences for customers.”
The NDP100 and NDP101 can be easily designed into battery-powered devices, including earbuds, headsets, remote controls, laptops and other voice-activated products, supporting entirely new form factors and usage models designed to wake up to speech rather than touch.
Key features of the NDP100:
● 1.4 x 1.8 mm 12 ball WLBGA
● Active power consumption of <150 µW while recognizing words
● Digital microphone interface or I2S streaming inputs
● 3 seconds of audio sample holding buffer
● Direct neural network access for sensor and other use cases
● Speech service for keyword training
● Device control software development kit (SDK) for seamless software integration
● Training development kit (TDK) for customer-programmed neural network applications
● General purpose Arm Cortex-M0 processor with 112KB RAM
Additional features of the NDP101:
● 5x5 mm QFN
● 8 GPIOs
● Serial flash boot option
● SPI master for sensor interfaces
Founded in 2017 and headquartered in Irvine, Calif., Syntiant Corp. is moving artificial intelligence and machine learning from the cloud to edge devices. Syntiant’s advanced chip solutions merge deep learning with semiconductor design to produce ultra-low-power, high performance, deep neural network processors for always-on applications in battery-powered devices, ranging from ear buds to smart speakers and mobile phones. Syntiant is backed by some of the world’s strongest strategic investors, including Intel Capital, Microsoft M12, Bosch Ventures and the Amazon Alexa Fund. More information can be found on www.syntiant.com or by following the company on Twitter @Syntiantcorp.
George Medici/Evan Pondel