| Paper: | SLP-P3.1 | ||
| Session: | Novel LVCSR Algorithms | ||
| Time: | Tuesday, May 16, 14:00 - 16:00 | ||
| Presentation: | Poster | ||
| Topic: | Speech and Spoken Language Processing: Alternative Statistical and Machine Learning Methods for General ASR (e.g., no-HMM methods) | ||
| Title: | Isolated-Word Recognition with Penalized Logistic Regression Machines | ||
| Authors: | Birkenes Oystein, Norwegian University of Science and Technology, Norway; Tomoko Matsui, Institute of Statistical Mathematics, Japan; Kunio Tanabe, Waseda University, Japan | ||
| Abstract: | We propose a new approach to isolated-word speech recognition based on penalized logistic regression machines (PLRMs). With this approach we combine the hidden Markov model (HMM) with multiclass logistic regression resulting in a powerful speech recognizer which provides us with the posterior probability for each word. Experiments on the English Eset show significant improvements compared to conventional HMM-based speech recognition. | ||
©2018 Conference Management Services, Inc. -||- email: webmaster@icassp2006.org -||- Last updated Friday, August 17, 2012