|
I have been working with speech technology for the past decade - both in academia and the industry.
My primary occupation is with Speech @ Microsoft where I am managing technology enhancement and innovation on the Core team. I'm also driving speech technology collaboration with Microsoft Research.
I received my PhD in Speech Technology from University College London. The topic of my dissertation is accent variation modeling and how to improve speech recognition accuracy for accented speakers. In short, my research offers a mechanism for automatically generating an idiodictionary, a user-specific pronunciation dictionary, at runtime based on accent features extracted directly from the user's speech.
I have been involved with the Department of Linguistics and the CLMS program for five years and I am teaching a course in Introduction to Speech Technology.
|