Built-in XMOS voice processor captures voice from afar and all directions
TOKYO (Dec. 3, 2018) – On Dec. 21, YUKAI Engineering Inc., a Tokyo-based robotics startup, and NTT DoCoMo, Japan’s largest mobile services provider, will jointly release codama, a voice interface development kit that can turn all kinds of electronic and IoT devices into voice-operable ones. Developers can integrate codama into their products to run them via NTT DoCoMo AI Agent API by using DoCoMo’s Raspberry Pi programming software. Featuring XMOS’s beamforming voice processor, codama enables devices to capture voice commands from distances, even in noisy environments, accurately interpret them, and respond.
The kit, which will be available on Amazon.co.jp and GitHub for 15,000 Japanese yen, will be demonstrated at DoCoMo Open House 2018 to be held at Tokyo Big Sight from Dec. 6-8, 2018.
Service- and Device-Agnostic Speech Interface
High-Performance Voice Recognition
codama uses XMOS XVF3100 that features:
Easy Steps to Create a Personalized Speech Dialogue Device
Users can go on YUKAI Engineering’s Website to create and register their own voice keywords for their devices. They may also opt to have codama devices respond to their personalized keywords only.
“Codama enables people to easily and inexpensively create their own, custom-designed voice-operable IoT devices. This is a game changer for businesses that are interested in tapping into the power of AI-supported speech communications to provide new services. This could not have been achieved without NTT DoCoMo’s forward vision to promote innovation through the initiative and XMOS’s cutting-edge technology,” YUKAI Engineering CEO Shunsuke Aoki said.
“NTT DoCoMo developed AI Agent API to ultimately provide end users with all-new speech AI experiences based on highly natural AI-supported communication. Today, we are very excited to introduce new speech development kit integrated with XMOS’s class-leading far-field voice interface technology working with Raspberry Pi and NTT DOCOMO AI Agent API seamlessly. This kit enables everybody to develop speech and dialog service “device” easily. We are very grateful to Yukai engineering and XMOS team collaboration to takes this major step forward for achieving our goal. We are expecting all users to enjoy new speech AI experiences,” Yoshikazu Akinaga, Manager of Innovation Management Department at NTT DoCoMo, said.
YUKAI Engineering is a robotics startup that develops and sells products befitting its slogan: “Make the world a fun place to live in with robotics technology.” YUKAI’s award-winning products include Necomimi, a headband with fake cat ears that move in sync with the user’s brain waves (selected for the Time magazine’s “The World’s 50 Best Innovations 2011” list), and BOCCO, a communication robot that keeps family members in touch with each other.
About NTT DoCoMo
NTT DoCoMo, Japan’s largest telecommunications company, provides innovative, convenient and secure mobile services that enable customers to realize smarter lives. A world-leading developer of 5G networks, NTT DoCoMo is also driving innovation in NFC infrastructure and services, emerging IoT solutions and many other mobile-related initiatives. Under its +d (pronounced plus d) initiatives, DOCOMO will create new added value with business partners to enrich people’s lives with all-new services and businesses.
XMOS is a leading supplier of voice interface solutions in the consumer electronics market. The company is an established brand in speech processing, biometrics and AI interface, which is achieved through a combination of XMOS’s unique silicon architecture and highly sophisticated software.