- 14:55 – 15:15
Improving the LLM experience with sensor provided context
Large Language Models (LLM’s) are a truly transformative technology that will alter many aspects of our lives. It has quickly become apparent that the usefulness of these LLMs is multiplied by the exactness of the prompt, leading to a rise in prompt engineering. In this presentation we will discuss how sensors on devices will play a significant role in helping generate more useful and contextual prompts and how audio and voice might play a role in the future of this technology.
Dr. Michael Pate
Dr. Michael Pate received advanced degrees from Texas Tech University and Denmark Technical University concentrating in the design of Class-D amplifiers for audio systems. After completing his Ph.D. work he began at Texas Instruments where he led the design of some of the highest power integrated Class-D amps on the market. From there Dr Pate moved to product definition roles at Cirrus Logic, Audience and Knowles Electronics where he focused on mixed signal audio, DSP and algorithms and MEMS microphones respectively. Having worked in all audio components as a semiconductor supplier he moved to Fitbit and now works at Google where he runs audio technology development across Pixel Phones, Pixel Buds and Pixel Watch.