I am a fifth year Ph.D. student in the Computer Science department at OSU. I am a member of the Slate Lab. I work on articulatory feature-based models for Automatic Speech Recognition. The main idea is to account for the large amounts of pronunciation variability in conversational speech by hypothesizing it as the result of asynchrony between the various articulators that are involved in speech production. My advisor is Prof. Eric Fosler-Lussier.

Publications

  1. BulletP. Jyothi, L. Johnson, C. Chelba, B. Strope. Large-scale Discriminative Language Model Reranking for Voice-Search. To appear in Proceedings of WLM (NAACL), 2012

  2. BulletP. Jyothi, L. Johnson, C. Chelba, B. Strope. Distributed Discriminative Language Models for Google Voice-Search. Proceedings of ICASSP, 2012

  3. BulletP. Jyothi, K. Livescu, and E. Fosler-Lussier. Lexical Access Experiments with Context-Dependent Articulatory Feature-Based Models. Proceedings of ICASSP, Prague, May 2011

  4. BulletP. Jyothi and E. Fosler-Lussier. Discriminative Language Modeling Using Simulated ASR Errors. Proceedings of Interspeech, Makuhari, September 2010

  5. BulletR. Prabhavalkar, P. Jyothi, W. Hartmann, J. Morris, E. Fosler-Lussier. Investigations into the Crandem Approach to Word Recognition. Proceedings of NAACL HLT, Los Angeles, June 2010

  6. BulletP. Jyothi, E. Fosler-Lussier. A Comparison of Audio-free Speech Recognition Error Prediction Methods. Proceedings of Interspeech, Brighton, September 2009

Contact

snail mail: 395 Dreese Laboratories, 2015 Neil Avenue, Columbus OH - 43201

email: last name (all letters in lower case)@cse.ohio-state.edu