top of page
Represents machine learning which is the technology behind our solution

Our Unique Approach to AI 

Advanced Model Architectures

Best in class, our advanced model architectures analyze both the semantic and acoustic aspects of speech - the words we say and how we say them. 

Superior performance, robustness and 
scalability

We employ the latest deep learning architectures for both natural language processing (NLP) and signal-based models, as well as best practices in testing and training of our models. In contrast to approaches that rely on feature design, our approach yields superior performance, robustness, and scalability.

Our models are speaker independent - no baseline required and are trained on diverse datasets. We don’t need to store any data for our models to optimally perform - maintaining data privacy and security. 

Technology

About Our AI

Advanced model architectures

The latest deep learning architectures for both NLP and signal-based models

Trained on large diverse data sets

Models work for unknown speakers from diverse backgrounds without requiring a baseline to additional information on the speaker

Academically validated

Our peer-reviewed papers follow best practices and appear in top speech technology publications

Rich information

We provide risk level estimates for depression and anxiety along with a range of analytics

Elizabeth Shriberg, PhD

Meet Our Chief Science Officer

Elizabeth Shriberg, PhD

​​Elizabeth Shriberg, PhD, is the Chief Science Officer and leads the machine learning team at Ellipsis Health. She is also a current Affiliate of Johns Hopkins University. Previously, Dr. Shriberg was a Senior Principal Scientist in Alexa AI at Amazon, working in speech and health. 

With two decades of experience as a Principal Scientist at SRI International, she led government and commercial R&D efforts in speech-based emotion and affective computing, health monitoring, speech understanding, dialog modeling, modeling of meetings and teams, computational prosody, speaker verification, deception detection, and modeling of speech disfluencies. She was also a Principal Researcher at Microsoft, where she developed methods for natural human-human-machine conversation. 

 

Dr. Shriberg has been an External Fellow of the International Computer Science Institute in Berkeley, CA. She has over 300 publications and patents, and received the International Speech Communication Association Fellow Award and the SRI International Fellow Award. Dr. Shriberg has served on a wide range of international academic association, conference, and journal boards. She also advises start-ups in the area of conversational AI.

Our Intellectual Property

bottom of page