Topic
Word error rate
About: Word error rate is a research topic. Over the lifetime, 11939 publications have been published within this topic receiving 298031 citations.
Papers published on a yearly basis
Papers
More filters
TL;DR: It is proposed that doubly confusable pairs, rather than high neighborhood densit y, may better explain phonetic neighborhood errors in human speech processing.
164 citations
TL;DR: In this article, a speech recognition system recognizes speech and generates one or more word strings, each of which is a hypothesis of the speech, and creates and stores a probability value or score for each of the word strings.
Abstract: The speech recognition system recognizes speech and generates one or more word strings, each of which is a hypothesis of the speech, and creates and stores a probability value (210) or score for each of the word strings. The speech recognition system also creates and stores, for each of the word strings, one or more keyword-value pairs (213) that represent semantic elements and semantic values of the semantic elements for the speech that was spoken. One or more dynamic semantic rules are defined that specify how a probability value of a word string should be modified based on the information about external conditions, facts, or the environment of the application in relation to the semantic values of the word string (214). The dynamic semantic rules are applied to word strings and keyword-value pairs (214). The speech recognizer modifies one or more probability values, re-orders the word strings (216), and returns control to the application.
164 citations
TL;DR: This paper deals with eigenchannel adaptation in more detail and includes its theoretical background and implementation issues, undermining a common myth that the more boxes in the scheme, the better the system.
Abstract: In this paper, several feature extraction and channel compensation techniques found in state-of-the-art speaker verification systems are analyzed and discussed. For the NIST SRE 2006 submission, cepstral mean subtraction, feature warping, RelAtive SpecTrAl (RASTA) filtering, heteroscedastic linear discriminant analysis (HLDA), feature mapping, and eigenchannel adaptation were incrementally added to minimize the system's error rate. This paper deals with eigenchannel adaptation in more detail and includes its theoretical background and implementation issues. The key part of the paper is, however, the post-evaluation analysis, undermining a common myth that ldquothe more boxes in the scheme, the better the system.rdquo All results are presented on NIST Speaker Recognition Evaluation (SRE) 2005 and 2006 data.
164 citations
TL;DR: This method of outlier handling combined with the classifier is applied to the well-known problem of automatic, constrained classification of chromosomes into their biological classes and it is shown that it decreases the error rate relative to the classical, normal, model by more than 50%.
164 citations
Patent•
11 Oct 2010
TL;DR: In this article, a method for identifying possible errors made by a speech recognition system without using a transcript of words input to the system is described. But this method does not consider the use of a word-to-word model.
Abstract: Methods are disclosed for identifying possible errors made by a speech recognition system without using a transcript of words input to the system. A method for model adaptation for a speech recognition system includes determining an error rate, corresponding to either recognition of instances of a word or recognition of instances of various words, without using a transcript of words input to the system. The method may further include adjusting an adaptation, of the model for the word or various models for the various words, based on the error rate. Apparatus are disclosed for identifying possible errors made by a speech recognition system without using a transcript of words input to the system. An apparatus for model adaptation for a speech recognition system includes a processor adapted to estimate an error rate, corresponding to either recognition of instances of a word or recognition of instances of various words, without using a transcript of words input to the system. The apparatus may further include a controller adapted to adjust an adaptation of the model for the word or various models for the various words, based on the error rate.
164 citations