This study focuses on word pronunciation in Text-to-Speech systems for Swedish. The purpose is to investigate whether machine learning techniques match knowledge-based systems in Swedish word pronunciation. The experiments show a maximum grapheme accuracy of just over 97%, and word accuracies from 67.0% for word pronunciation excluding stress assignment, which compares favourably to existing kn...

We propose a method for modeling pronunciation variation in the context of spell checking for non-native writers of English. Spell checkers, typically developed for native speakers, fail to address many of the types of spelling errors peculiar to non-native speakers, especially those errors influenced by differences in phonology. Our model of pronunciation variation is used to extend a pronounc...

Recent research has shown that a properly designed ASRbased CALL system (Dutch-CAPT) was capable of detecting pronunciation errors and of providing comprehensible feedback on pronunciation. Since pronunciation is not the only skill required for speaking a second language, we explored the possibility of extending the Dutch-CAPT approach to other aspects of speaking proficiency like morphology an...

Mispronunciation is still among the frailties of language learners the world over. The present study reports an all-out investigation into pronunciation training for the Persian-speaking EFL learners at the intermediate level of language proficiency. More specifically, the present study investigated (a) the underlying causes of mispronunciation at the level of segmental features, (b) the crucia...

Introducing pronunciation models into decoding has proven beneficial for LVCSR. As Minimum Phone Error (MPE) training has almost become a standard scheme for acoustic modeling, a discriminative pronunciation modeling method is investigated under the framework of MPE training. In order to bring the pronunciation models into MPE training, the auxiliary function of MPE training is rewritten at wor...

This paper discusses several relevant issues concerning the pronunciation of proper names. Although it was motivated by the experience of the ONOMASTICA European project, it re ects a personal view, not constituting, therefore, an o cial document of the project. The major outcome of the project was the production of two important linguistic resources: the set of 11 national pronunciation lexica...

This paper presents an approach for the generation and selection of pronunciation transcriptions for a exible word recognizer. The basic idea is to produce pronunciation variants and corresponding scores with a set of pronunciation variation rules, which are weighted with their frequencies of occurence measured on the training data. This approach addresses the problem of interfering transcripti...

In this research, we propose a hybrid approach for acoustic and pronunciation modeling for Arabic speech recognition. The hybrid approach benefits from both vocalized and non-vocalized Arabic resources, based on the fact that the amount of non-vocalized resources is always higher than vocalized resources. Two speech recognition baseline systems were built: phonemic and graphemic. The two baseli...

Recently, a novel acoustic representation of speech was proposed, where dimensions of the non-linguistic factors can hardly be seen. Using this structural representation, individual learners were described as distorted phonemic structures and automatic evaluation of the pronunciation was investigated. This paper describes two new analyses using the proposed method. The first analysis is done to...

This paper discusses several relevant issues concerning the pronunciation of proper names. Although it was motivated by the experience of the ONOMASTICA European project, it reeects a personal view, not constituting, therefore, an oocial document of the project. The major outcome of the project was the production of two important linguistic resources: the set of 11 national pronunciation lexica...