I am now working as a Research Scientist with the Amazon Alexa team in Cambridge, Massachusetts.
I was awarded my B.Tech. and Ph.D. degrees in Electrical Engineering from the Indian Institute of Technology (IIT) Kanpur, India.
My Ph.D. thesis was entitled "Analysis of Pitched Polyphonic Music for Source Transcription", reflecting my research interests in acoustic signal analysis and music information retrieval.
Whilst working in the Language and Brain lab, University of Oxford, I helped to develop the FlexSR automatic speech recognition (ASR) technologies using phonological considerations within the framework of the Featurally Underspecified Lexicon (FUL) model:
- A phonological feature based transfer learning scheme for creating ASR systems for under-resourced languages
- A phonological feature based computer-aided pronunciation training system that detects mispronunciations and supplies corrective feedbacks to the learners of a second language
Vipul Arora, Laxmidhar Behera, Tharun Reddy and Ajay Pratap Yadav, "HJB Equation Based Learning Scheme for Neural Networks", in International Joint Conference on Neural Networks (IJCNN), May 2017.
Vipul Arora, Aditi Lahiri and Henning Reetz, "Automatic Speech Recognition: what phonology can offer" in Lexical Processing of Speech, deGruyter/Mouton, 2017.
Vipul Arora, Aditi Lahiri and Henning Reetz, "Attribute based shared hidden layers for cross-language knowledge transfer", in IEEE Workshop on Spoken Language Technology (SLT), San Diego, California, December 2016.