Context-dependent ASR
نویسنده
چکیده
Computer speech recognition gains more and more attention these days with its implementation in nearly everyday life. But the ultimate goal is still out of reach. The automatic recognition (ASR) systems can very precisely work on small domain. However the bigger the domain is the worse is the performance of the ASR system. The aim of many researchers is to diminish this problem on various levels of the ASR. This work describes components of an ASR system, how they are working together and delves into prosody and how it is used in ASR. From the usage of prosody, the main part of work describes how the ASR can be improved better modeling of the speech variance. We discuss usage of triphones, syllables and other models as well as algorithms and techniques for clustering. Copies of this report are available on http://www.kiv.zcu.cz/publications/ or by surface mail on request sent to the following address: University of West Bohemia in Pilsen Department of Computer Science and Engineering Univerzitn¶3 8 30614 Pilsen Czech Republic Copyright © 2009 University of West Bohemia in Pilsen, Czech Republic
منابع مشابه
Comparison of acoustic modeling techniques for Vietnamese and Khmer ASR
This paper presents a comparison of some different acoustic modeling strategies for under-resourced languages. When only limited speech data are available for under-resourced languages, we propose some crosslingual acoustic modeling techniques. We apply and compare these techniques in Vietnamese ASR. Since there is no pronunciation dictionary for some underresourced languages, we investigate gr...
متن کاملPerformance Improvement of Dysarthric Speech Recognition Using Context-Dependent Pronunciation Variation Modeling Based on Kullback-Leibler Distance
In this paper, we propose context-dependent pronunciation variation modeling based on the Kullback-Leibler (KL) distance for improving the performance of dysarthric automatic speech recognition (ASR). To this end, we construct a triphone confusion matrix based on KL distances between triphone models, and build a weighted finite state transducer (WFST) from the triphone confusion matrix. Then, d...
متن کاملMulti-level acoustic modeling for automatic speech recognition
Context-dependent acoustic modeling is commonly used in large-vocabulary Automatic Speech Recognition (ASR) systems as a way to model coarticulatory variations that occur during speech production. Typically, the local phoneme context is used as a means to define context-dependent units. Because the number of possible context-dependent units can grow exponentially with the length of the contexts...
متن کاملA study of implicit and explicit modeling of coarticulation and pronunciation variation
In this paper, we focus on the modeling of coarticulation and pronunciation variation in Automatic Speech Recognition systems (ASR). Most ASR systems explicitly describe these production phenomena through context-dependent phoneme models and multiple pronunciation lexicons. Here, we explore the potential benefit of using feature spaces covering longer time segments in terms of implicit modeling...
متن کاملPronunciation Variations and Context-dependent Model to Improve ASR Performance for Dyslexic Children’s Read Speech
Focusing on the key element for an ASR-based application for dyslexic children reading isolated words in Bahasa Melayu, this paper can be an evidence of the need to have a carefully designed acoustic model for a satisfying recognition accuracy of 79.17% on test dataset. Pronunciation variations and context-dependent model are two main components of such acoustic model. This model adopts the mos...
متن کاملOn recognition of non-native speech using probabilistic lexical model
Despite various advances in automatic speech recognition (ASR) technology, recognition of speech uttered by non-native speakers is still a challenging problem. In this paper, we investigate the role of different factors such as type of lexical model and choice of acoustic units in recognition of speech uttered by non-native speakers. More precisely, we investigate the influence of the probabili...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2009