(Enter summary)
Abstract: In this paper we define two alternatives to the familiar perplexity
statistic (hereafter lexical perplexity), which is widely applied both
as a measure-of-goodness and as an objective function for training
languagemodels. These alternatives, respectively acoustic perplexity
and the synthetic acoustic word error rate, fuse information
from both the language model and the acoustic model. We show
how to compute these statistics by effectively synthesizing a large
acoustic corpus, demonstrate... (Update)
Context of citations to this paper: More
...increasingly investigated lately, with a special focus on how to estimate a Word Error Rate on test data without having to decode them. We are currently investigating the use of the the Synthetic Acoustic Word Error Rate (SAWER) introduced in [10], which fuses information from...
Cited by: More
Predicting Word Correct Rate from Acoustic and Linguistic - Confusability Gies Bouwman
(Correct)
Automatic Generation and Selection of Multiple.. - Deligne, Maison.. (2001)
(Correct)
Automatic Generation And Selection Of Multiple.. - Deligne, Maison.. (2001)
(Correct)
Active bibliography (related documents): More All
0.5: Construction of Corpus-Based Syntactic Rules for Accurate.. - Itosaka, Takezawa (1992)
(Correct)
0.2: A Probabilistic Approach to the Problem of Automatic Selection .. - Chuang, Hwang (1996)
(Correct)
0.2: Randomized Algorithms - A short course on and the probabilistic.. - Merkle (2001)
(Correct)
Similar documents based on text: More All
0.5: Topic Change And Local Perplexity In Spoken Legal Dialogue - Kenne, O'Kane
(Correct)
0.3: Evaluation Metrics For Language Models - Chen, Beeferman, Rosenfeld (1998)
(Correct)
0.3: Analyzing and Predicting Language Model Improvements - Iyer, Ostendorf, Meteer (1997)
(Correct)
Related documents from co-citation: More All
2: Automatic phonetic baseform determination (context) - Bahl, Das et al. - 1991
2: Modeling pronunciation variation for ASR: A survey of the literature (context) - Strik, Cucchiarini - 1999
2: A User-Configurable System for Voice Label Recognition
- Rose - 1996
BibTeX entry: (Update)
H. Printz and P. Olsen, "Theory and practice of acoustic confusability", ASR 2000, pp 77-84. http://citeseer.ist.psu.edu/printz00theory.html More
@misc{ printz00theory,
author = "H. Printz and P. Olsen",
title = "Theory and practice of acoustic confusability",
text = "H. Printz and P. Olsen, Theory and practice of acoustic confusability,
ASR 2000, pp 77-84.",
year = "2000",
url = "citeseer.ist.psu.edu/printz00theory.html" }
Citations (may not include all citations):
202
Statistical Methods for Speech Recognition (context) - Jelinek - 1997
65
Mathematical Analysis (context) - Apostol - 1957
30
Introduction to Mathematical Statistics (context) - Hoel - 1984
29
An inequality for rational functions with applications to so.. (context) - Gopalakrishnan, Kanevsky et al. - 1991
22
Evaluation metrics for language models
- Chen, Beeferman et al. - 1998
18
Fabricating conversational speech data with acoustic models:.. (context) - McAllaster, Gillick et al. - 1998
9
Perplexity - a measure of the difficulty of speech recogniti.. (context) - Bahl, Baker et al. - 1977
6
Analyzing and predicting language model improvements
- Iyer, Ostendorf et al. - 1997
5
The applicability of adaptive language modelling for the bro..
- Clarkson, Robinson - 1998
4
Elementary Probability Theory with Stochastic Processes (context) - Chung - 1979
2
Measuring information provided by language model and acousti.. (context) - Ferretti, Maltese et al. - 1990
1
A new metric for stochastic language model evaluation (context) - Ito, Kohda et al. - 1999
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