The industry is still trying to absorb the full impact of Apple's announcements during its annual WorldWide Developers Conference (WWDC) 2021. In particular the confirmation that Apple is bringing Siri to third-party devices and platforms, allowing HomeKit-supporting companies to integrate the digital assistant into their devices. This support is expected to become a reality in the market after Fall 2021, and third-party devices should follow quickly before the end of the year. With the third-party support, HomeKit accessory makers will be able to integrate Siri into their own devices.
During WWDC21, Apple demoed this on stage by querying Siri on an Ecobee Smart Learning Thermostat, the first officially confirmed device with this integration. Apple also confirmed that Siri support on third-party devices will enable features such as Personal Requests, Intercom, timers, and alarms. Hardware developers must be a part of Apple's MFi program for HomeKit accessories, and users will still need to have an Apple device, such as an Apple Watch or HomePod mini to interface using voice.
Sensory is one of the first voice technology providers to confirm development support for this integration. The Californian company announced that TrulyHandsfree, the company’s edge-based wake word and phrase recognition engine, now supports models for ‘Hey Siri’ for dozens of different countries.
Developers licensing Sensory’s TrulyHandsfree SDK can request access to the Hey Siri models for integration into their prototypes and finished products. Sensory’s Hey Siri wake word has already been used in products that have received MFi (Made for iPhone/iPad) approval, now Sensory is releasing international versions, and the timing is perfect for Apple’s introduction of Hey Siri for third-party devices through HomeKit and Matter as announced in Ecobee with Hey Siri.
Sensory is recognized as the de-facto leader for custom branded wake words and earlier this year the company opened its online portal, VoiceHub, to developers for quick development of custom wake words. While demand for custom branded wake words continues to be strong, there is also a robust demand for tuned wake word models that support well-known virtual assistants like Alexa, Bixby, Cortana, Google, and Siri. With the release of the Hey Siri model, the company now supports all of the most popular virtual assistants in the USA. Looking beyond the USA, Sensory also provides wake word models for Tencent, Baidu, Naver, Rakuten, and others. The new Siri models support multiple languages across North America, Europe, and Asia.
“Every conversation with a virtual assistant, starts with a wake word, so accurate wake word performance across the world is mandatory for a positive user experience,” says Todd Mozer, Sensory CEO. “Sensory’s TrulyHandsfree wake word engine has shipped in billions of products, including mobile phones, smart speakers, wearables, smart appliances, and automobiles. All of that real-world experience provides a competitive advantage for Sensory models. Our wake words provide unparalleled accuracy with the acoustic challenges of background noise, accented speech, and far-field conditions.”
Sensory provides developers with a preset library of popular tuned wake word models, support for two dozen languages/dialects, as well as the ability to create custom-branded wake words with VoiceHub. As a member of Amazon’s Voice Interoperability Initiative, Sensory is well-positioned to support the mission of providing customers choice and flexibility to interact with multiple voice services. Third-party device makers can create new products that integrate multiple virtual assistants with branded wake words and custom voice experiences. As stated by Jeff Bezos, Amazon CEO, “Multiple simultaneous wake words provide the best option for customers.”
www.sensory.com
