Senior Research Scientist, 5/2018 – present
Pryon is an AI company that connects enterprise employees to digital transformation through an Augmented Intelligence platform, extending their ability to find and use knowledge, drive workflows, and make better decisions from wherever they are. My roles have included:
- Developing context based interactions in an IR-QA system
- Conducting deep dive analytics to direct efforts in both engineering and machine learning
- Managing interns in exploratory projects in analyzing question type, using clustering for identifying frequent questions
- Designing and managing annotation projects from QA examples and named entity recognition to visual documentation segmentation
Adjunct associate Professor, 5/2018 – 7/2020, Computer Science.
- Teaching graduate courses in Computational Linguistics and Speech Recognition and working with companies and students as the Industry Liaison for our CL students.
Associate Professor of Practice, 7/2016 – 5/2018, Computer Science.
- Teaching graduate courses and Industry Liaison in the Computational Linguistics MA program.
Teaching: Graduate courses in Computational Linguistics. Courses include:
- CS136 Automated Speech Recognition. This course covers speech recognition from both applied and theoretical perspectives.
- CS140 Natural Language Annotation for Machine Learning . Experiential learning where students work in groups to create a training corpus for machine learning and create and execute an annotation task.
- CS233-234: Discourse and Dialog. Seminar doing a deep dive into both discourse analysis and the dialog systems that now populate our world, such as Siri & Alexa.
- Summer intensive program in Mobile Voice Applications, Summers 2014-16. 7 week program where students learn the skills and in groups build an interactive voice application.
- Organized “Industry Reception” bringing ~15 companies in speech and NLP to Brandeis to meet our graduate students in the CL program.
- Individually advised students on their resumes, interviewing, and negotiating offers.
Adjunct Professor, 1/2010 – 7/2016. Computer Science.
- Teaching graduate courses in Computational Linguistics and Speech Recognition. Created the Industry Liaison role, bringing companies to Brandeis to meet our students
Working since 2009 with organizations to help them employ cutting edge technologies in information extraction, machine learning, and speech recognition. Current projects include:
- McLean Hospital: Consultant on NIH funded project using machine learning for information extraction from medical health records of psychiatric patients to improve prediction of outcome. Supervising Brandeis student intern.
- Partner’s Healthcare: Consultant on Brigham and Womens’ Hospital AHRQ project “NLP to Improve Accuracy and Quality of Dictated Medical Documents” (Grant number R01 HS024264).
- BBN Avoke: Consultant on Call Center Analytics projects, developing statistical language models for multiple languages (English, European and Asian languages) and using unsupervised and semi-supervised topic modeling to track important “episodes” in a call.
- NavCanada: Provide technical support on the use of speech recognition in Air Traffic Control training and simulation software and help design training materials for instructors and users.
Other Clients include: CallMiner, Ditech Networks, General Catalyst, RAMP Holdings, Speech Morphing, Transcend, UFA, US Department of Justice, Vlingo, Xtone.Consulting Engagements
Expert counsel and evaluation of market, business, and technology decisions related to speech and natural language domains. All engagements are developed with the client and tailored to specific needs.
|Description||Examples from previous engagements|
|Technology Direction||Working with employees and interns on a project where I bring specific expertise to guide the work.||Information extraction from healthcare records, topic identification in call center analytics, language modeling for speech recognition|
|Market/ Technology Evaluations||Ranging from evaluations of a company prior to an investment decision to evaluation of a product’s performance on specific data.||Speech recognition performance, information extraction and topic detection, and speech analytics|
|Market/ Technology Research||Targeted at collecting and analyzing literature on a paritucular technology or set of technologies and includes comparisons of specific products and analysis of the latest research in that area.||Voicemail to text, medical transcription, call center analytics|
|Seminars and Workshops||Designed to provide both a broad view of the speech recognition or natural language processing technologies in your industry and an in depth view of the particular aspects that directly impact your business.||Medical Transcription, Closed Captioning, Air Traffic Control Training, and Opinion Research|
|Patent Research||Finding prior art related to a particular patent, including development of claims charts. In Speech Recognition, the work takes advantage of SpeechMap, which is a collection of “hard to find” materials dating back to the 1980’s and 1990’s.||Dictation, interactive systems, core speech algorithms|
Vice President (2007-2009) Speech and Content.
- Responsible for the strategic direction of speech and natural language processing technologies to provide search and media marketing for customers such as Fox, Boston.com, NBC, and the Patriots. Analysis of customer data to determine the best strategy for search engine optimization and content merchandizing. Managed patent portfolio, identifying patentable technologies and working with the legal team on new patents.
Vice President (6/04 – 4/07) Commercial Speech Solutions, Delta Division.
- Responsible for technology direction of speech products in both network telephony and large vocabulary speech recognition for search. Product and market evangelism, pre-sales support, product definition, patent portfolio, conduit to BBN R&D team for new technology.
Department Manager (6/03 – 6/04)
- Call Center Speech Solutions Department. Responsible for a new department focused on products aimed at the telephony speech market. Achieved significant year over year growth and profitability.
Director (1/00 – 6/03)
- Program Director for new applications of speech and language technologies in the Customer Contact Center market; business development; negotiating partnerships and VAR agreements; managing the development and delivery of products and services, business planning.
R&D Manager (8/98 – 1/00) Speech and Language Processing Department.
- Directed more than 50 researchers in speech recognition, topic identification, telephone dialog systems, and information extraction from text and speech. Responsible for line management of $10M in revenue.
Senior Scientist (8/93- 8/98) Speech and Language Processing Department.
- Co-PI of the Large Vocabulary Speech Recognition project, a $2M per year effort to improve speech recognition performance.
Scientist (8/87 – 9/93) Artificial Intelligence Department.
- Primary developer of the SPOKESMAN Generation System, which produced text for several applications, including DARPA’s AirLand Battle Management Project.