Stefan Ortmanns, Hermann Ney, Thorsten Firzlaff  Home/Search
Context Related
| informatik.rwthaa...OSPEECH97_Sep97.ps Cached: PS.gz PS PDF This document uses CoBlitz to cache paper downloads. If your firewall is blocking outgoing connections to port 3125, you can use these links to download local copies.
Image Update HelpPS.gz PS PDF From: i6.informatik.r...blikationen1997 (more) (Enter author homepages) |
Rate this article:      (best)Comment on this article |
Abstract: This paper studies algorithms for reducing the computational effort of the mixture density calculations in HMM-based speech recognition systems. These likelihood calculations take about 70 \Gamma 85% of the total recognition time in the RWTH system for large vocabulary continuous speech recognition. To reduce the computational cost of the likelihood calculations, we investigate several space partitioning methods. A detailed comparison of these techniques is given on the North American Business... (Update)
Context of citations to this paper: More
.... in our decoder: language model lookahead [7] phoneme look ahead [8] and fast likelihood calculation of continuous mixture densities [9][4] 3.1. Enhanced Language Model Look Ahead For more ecient pruning during beam search, we smear the language model probabilities...
...mixture densities. 2. 1 Hypothesis Pruning During search we use conventional beam pruning and histogram pruning (e.g. Ortmanns et al. 1997a) at state and word level. We found that especially histogram pruning at the word level was capable to reduce the computational effort...
Cited by: More
State-Based Gaussian Selection In Large Vocabulary.. - Gales, Knill, Young (1998) (Correct)
Fast Search for Large Vocabulary Speech Recognition - Kanthak, Sixtus, Molau.. (2000) (Correct)
Recent Improvements Of The RWTH Large Vocabulary.. - Sixtus, Molau.. (2000) (Correct)
Active bibliography (related documents): More All
0.7: Using SIMD Instructions For Fast Likelihood Calculation In.. - Kanthak, Schütz, Ney (2000) (Correct)
0.7: Look-Ahead Techniques For Improved Beam Search - Ortmanns, Ney, Eiden, Coenen (1996) (Correct)
0.6: Architecture and Search Organization for Large.. - Ortmanns.. (Correct)
Similar documents based on text: More All
0.5: The RWTH Large Vocabulary Speech Recognition.. - Kanthak, Molau.. (2000) (Correct)
0.5: Within-Word vs. Across-Word Decoding for Online Speech.. - Kanthak, Sixtus, Molau.. (2000) (Correct)
0.3: Statistical Image Object Recognition using Mixture Densities - Dahmen, Keysers, Ney, Güld (2000) (Correct)
BibTeX entry: (Update)
S. Ortmanns, H. Ney, T. Firzla : Fast Likelihood Computation Methods for Continuous Mixture Densities in Large Vocabulary Speech Recognition, Proc. European Conf. on Speech Communication and Technology, pp. 139-142, Rhodos, Greece, September 1997. http://citeseer.ist.psu.edu/ortmanns97fast.html More
@inproceedings{ ortmanns97fast,
author = "Stefan Ortmanns and Thorsten Firzlaff and Hermann Ney",
title = "Fast Likelihood Computation Methods for Continuous Mixture Densities in Large Vocabulary Speech Recognition",
booktitle = "Proc. Eurospeech '97",
address = "Rhodes, Greece",
pages = "139--142",
year = "1997",
url = "citeseer.ist.psu.edu/ortmanns97fast.html" }
Citations (may not include all citations):22 Communications of the ACM (context) - Bentley, Search et al. - 1975
8 Algorithm for Fast Approximative Evaluation of Diagonal Mixt.. (context) - Fritsch, Rogina et al. - 1996
5 Coenen: LookAhead Techniques for Fast Beam Search (context) - Ortmanns, Eiden et al. - 1997
5 Ney: Continuous Speech Recognition Tests and Results for the.. (context) - Dugast, Kneser et al. - 1995
4 Ney: Implementation of Word Based Statistical Language Model.. (context) - Wessel, Ortmanns - 1997
3 Finkel: An Algorithm for Finding Best Matches in Logarithmic.. (context) - Friedman, Bentley - 1977
2 Speech and Signal Processing (context) - Bocchieri, for et al. - 1993
2 Young: Use of Gaussian Selection in Large Vocabulary Continu.. (context) - Knill, Gales - 1996
2 Fifth European Signal Processing Conference (context) - Ney, of et al. - 1990
1 Ullrich: Hamming Distance Approximation for a Fast Log-Likel.. (context) - Beyerlein - 1995
1 Ullrich: Personal Communication (context) - Beyerlein - 1997
1 Nayar: Closest Point Search in High Dimensions (context) - Nene - 1996
Documents on the same site (http://www-i6.informatik.rwth-aachen.de/HTML/Publikationen/Publikationen1997.html): More
Acoustic Front-End Optimization for Large Vocabulary.. - Welling, Haberland, Ney (Correct)
Adaptive Topic-Dependent Language Modelling Using.. - Martin, Liermann, Ney (1997) (Correct)
A DP based Search Using Monotone Alignments in.. - Tillmann, Vogel, Ney, .. (1997) (Correct)
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC