SLS TECHNOLOGIES

Speech Recognition

SUMMIT is a speech recognition system developed by the SLS group. This speech recognizer uses a landmark-based approach for modeling acoustic-phonetic events and utilizes finite-state transducer (FST) technology to efficiently represent all aspects of the speech hierarchy including the phonological rules, the lexicon, and the probabilistic language model.

References

J. Glass, " A Probabilistic Framework for Segment-Based Speech Recognition," Computer Speech and Language 17 (2003), pp. 137-152. (PDF)

Hans J.G.A. Dolfing and I. Lee Hetherington, "Incremental Language Models for Speech Recognition using Finite-State Transducers," Proc. IEEE Automatic Speech Recognition and Understanding Workshop, Madonna di Campiglio, Italy, December 2001. (PDF)

L. Hetherington, "An Efficient Implementation of Phonological Rules using Finite-State Transducers," Proc. Eurospeech 2001, Aalborg, Denmark, September 2001. (PDF)

T. J. Hazen, I. Lee Hetherington and A. Park, "FST-Based Recognition Techniques for Multi-Lingual and Multi-Domain Spontaneous Speech," Proc. Eurospeech 2001, Aalborg, Denmark, September 2001. (PDF)

N. Ström, L. Hetherington, T.J. Hazen, E. Sandness and J. Glass, "Acoustic Modeling Improvements in a Segment-Based Speech Recognizer," Proc. 1999 IEEE ASRU Workshop, Keystone, CO, December 1999. (PDF)

J. Glass, T.J. Hazen and L. Hetherington, "Real-time telephone-based speech recognition in the JUPITER domain," Proc. ICASSP 99, Phoenix, AZ, March 1999. (PDF)