
XMOS boosts AI wake word microcontrollers
XMOS has launched two voice processors with machine learning wireless speakers, TVs, set-top boxes, smart home appliances and gateway products with a version that embeds Amazon Alexa wake word detection.
The XVF3610 and XVF3615 voice processors are built on the xcore.ai architecture and will be used in a new reference platform. The designs mark an evolution of the XVF3510 voice processor already used in Amazon Alexa systems, adding audio echo cancellation (AEC) and interference cancellation algorithms to suppress noise from dominant sources and enable cross-room barge-in. The XVF3610 and XVF3615 also include automatic audio reference delay calibration – a critical capability for voice enabled TV systems – expanding the range of devices and environments where voice interfaces with wake word detection.
The XVF3615 adds embedded Amazon wake word for single pass Alexa enabled systems, allowing manufacturers to offload wake word processing from their host SoC systems. The microcontoroller will be available in distribution in January 2022 but is being used for the alpha programme of a new customisable voice reference design for voice, wakeword, speech-to-intent and intelligent voice applications called Avona. This allows designers to incorporate keyword, event detection or advanced local dictionary support to create a complete voice interface.
Developers can use the xcore software development kit (SDK) with software in C or C++ on a FreeRTOS environment for user application code for both bare-metal and RTOS use. This provides the ability to add bespoke differentiation for the end product. The reference design will be generally available in early 2022.
“The IoT market is creating new verticals and opportunities at breakneck speed, and we are seeing significant demand within the smart home. As more applications become commercially attractive or viable, engineers and designers need to be able to rely on hardware that offers agility and cost-effective performance without compromise on quality,” said Mark Lippett, CEO of XMOS.
“Versatility is the key with all three of these new voice solutions, inherent in the structure of the xcore.ai chip. Designers are empowered to make choices: what balance of compute do we want? How much of the voice element is customisable? Are we coding on bare metal or in FreeRTOS? That dexterity will help them innovate in ways that they otherwise couldn’t,” said Rob Fisher, Product Director at XMOS,
The XVF3610 is available now in distribution on standard lead times.
Related articles
- XMOS launches SDK for the ‘artificial intelligence of things’
- XMOS, AWS team for symmetric multiprocessing FreeRTOS
- XMOS and Plumerai partner on binarised neural networks
- XMOS acquires Setem Technologies for voice AI
- Infineon backs XMOS to work where audio meets AI
Other articles on eeNews Europe
- Nano Dimension launches DragonFly IV 3D-printer, acquires Swiss PCB maker
- Lattice snaps up its AI software provider
- ASML visit to boost European CHIPS Act
- High precision metrology for 3D heterogeneous packaging
- Xilinx launches its most powerful FPGA card
- Major security vulnerabilities found in DRAM
- Intel celebrates 50 years of the 4004
- Toshiba splits into three
