Official media partner
Events
Product Development Symposium
Interactive Voice and DSP
Session
Thursday 15 – 3:00pm – 4:00pm EDT
One 50 Minute Session followed by Q&A
Everything You Wanted to Know About Wake Words
Presenter: Jeff Rogers, Co-Founder of Sensory, Portland, Oregon USA
Content
Wake words have become common place in many markets and industries today. From one of the first always on and listening mobile phones (Moto X with Sensory TrulyHandsfree) to billions of mobile phones (many that use Sensory) to smart speakers, automotive, appliances, applications, medical devices, sports cameras, wearables and more, wake words are now a part of everyday life for many.
In this workshop, we will discuss the following:
Recommendations and rules on how to choose a good wake word. (Linguistic guidelines for good wake word performance)
Overview of False Reject/False Accept relation and balancing between the two for optimal performance
Model size vs. performance
Different types of wake words, their applications and strengths/weaknesses of each type
Fixed trigger
Enrolled fixed trigger
User defined trigger
Continuous adaptation for wake words
Speaker verification and speaker ID for wake words
Privacy, latency and other concerns
Demo tool on creating fast and free prototype wake words via new Sensory VoiceHub tool!
This workshop is targeted at product managers, product development teams, engineering managers, development engineers, product marketing and new ideation managers and teams.
Jeff Rogers
Co-Founder of Sensory, Portland, Oregon USA
Jeff Rogers serves as the Vice President of Sales of Sensory, the company recognized for being the originator of wake words for personal voice assistants. Sensory is a privately held company focused on improving user experiences through embedded machine learning technologies such as voice and natural language processing. As a Co-Founder of Sensory, Jeff has been involved and influential in the development and design on 100’s of products over the last 26 years that use speech and computer vision technologies. In the early years, Jeff was focused on selling Sensory semiconductors that included many speech technologies. As Sensory morphed into a software and IP licensing company, Jeff changed focus to licensing and working not only with large OEMs but also many chip companies and IP partners such as DSP Concepts. Jeff has licensed large companies such as Amazon, Google, Samsung, Microsoft, Huawei, Lenovo and many more.
Headquartered in Santa Clara, Calif., DSP Concepts is the leading provider of audio development tools and IP for creators of audio and voice-enabled products. Its Audio Weaver platform allows developers to easily test and deploy new features through a graphical drag-and-drop interface, and is currently powering over 40M of the world’s best devices for top brands in automotive and consumer products, including Tesla, GoPro, Braun, and Porsche.
The Leading Global Audio Engineering & Development Magazine
audioXpress magazine is the reference information source about innovative solutions and technologies in audio electronics, acoustic and electro-acoustic design for developers, audio product managers, the industry and its channel partners. Our increased focus on audio product development and design, particularly in cutting-edge growth segments such as voice recognition, hearables, sensors, and wireless and streaming audio technologies, reflect the evolving trends of the audio industry itself. Together with Voice Coil, the periodical for the loudspeaker industry, Circuit Cellar, the world’s source for embedded electronics engineering, and our Market Directories, we excel in information for product development and the product design communities.