Non-native Pronunciation Modeling in a Command & Control Recognition Task: A Comparison between Acoustic and Lexical Modeling
نویسنده
چکیده
In order to improve automatic recognition of English commands spoken by non-native speakers, we have modeled non-native pronunciation variation of Dutch, French and Italian. The results of lexical and acoustical modeling appeared to be source language and speaker dependent. Lexical modeling only resulted in a substantial improvement (of 35%) for the French speakers. Acoustic model adaptation halved the word error rates for the Italian speakers, whereas no improvements were found by lexical modeling of frequently observed Italian-accented non-native pronunciation variants. The performance for the Dutch speakers only slightly improved by lexical and acoustic modeling.
منابع مشابه
Integration of MLLR adaptation with pronunciation proficiency adaptation for non-native speech recognition
To recognize non-native speech, larger acoustic/linguistic distortions must be handled adequately in acoustic modeling, language modeling, lexical modeling, and/or decoding strategy. In this paper, a novel method to enhance MLLR adaptation of acoustic models for non-native speech recognition is proposed. In the case of native speech recognition, MLLR speaker adaptation was successfully introduc...
متن کاملTransfer Learning based Non-native Acoustic Modeling for Pronunciation Error Detection
The scarcity of large-scale non-native corpora and human annotations are two fundamental challenges in the development of computer-assisted pronunciation training (CAPT) systems. We explored several transfer learning based methods to detect the pronunciation errors without using nonnative training data. Effects were confirmed in the Mandarin Chinese pronunciation error detection of Japanese spe...
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 کاملNon-native Pronunciation Variation Modeling for Automatic Speech Recognition
Communication using speech is inherently natural, with this ability of communication unconsciously acquired in a step-by-step manner throughout life. In order to explore the benefits of speech communication in devices, there have been many research works performed over the past several decades. As a result, automatic speech recognition (ASR) systems have been deployed in a range of applications...
متن کاملFrame-Level Selective Decoding Using Native and Non-native Acoustic Models for Robust Speech Recognition to Native and Non-native Speech
v Regarded as a mismatch problem between the training and test conditions § Training condition: native speech § Testing condition: non-native speech § Widely used methods in speaker or environment adaptation v Research works dedicated to non-native ASR § Acoustic modeling § Pronunciation modeling § Language modeling § Hybrid modeling § Many researches uses a small amount of non-native speech Wh...
متن کامل