Phoneme Set Design Considering Integrated Acoustic and Linguistic Features of Second Language Speech
نویسندگان
چکیده
Recognition of second language speech is still a challenging task even for state-of-the-art automatic speech recognition (ASR) systems. Considering that second language speech usually includes less fluent pronunciation and mispronunciation even when it is grammatically correct, we propose a novel phonetic decision tree (PDT) method considering integrated acoustic and linguistic features to derive the phoneme set for second language speech recognition. We verify the efficacy of the proposed method using second language speech collected with a translation game type dialogue-based English CALL system. Experimental results demonstrated that the derived phoneme set achieved higher accuracy recognition performance than the canonical one.
منابع مشابه
Allophone-based acoustic modeling for Persian phoneme recognition
Phoneme recognition is one of the fundamental phases of automatic speech recognition. Coarticulation which refers to the integration of sounds, is one of the important obstacles in phoneme recognition. In other words, each phone is influenced and changed by the characteristics of its neighbor phones, and coarticulation is responsible for most of these changes. The idea of modeling the effects o...
متن کاملA Study of the Relationship between Acoustic Features of “bæle” and the Paralinguistic Information
Language users benefit from special phonetic tools in order to communicate linguistic information as well as different emotional aspects and paralinguistic information through daily conversation. Having functions in conveying semantic information to listeners, prosodic features form the essential part of linguistic behavour, manipulating them potentially can play an important role in transmitt...
متن کاملطراحی الگوریتم بازشناسی واجها با به کارگیری همبسته های آکوستیکی مشخصه های واجی
In the present paper, the phonological feature geometry of the Persian phonemes is analyzed in the form of articulate-free and articulate-bound features based on the articulator model of the nonlinear phonology. Then, the reference phonetic pattern of each feature that consists of one or a set of acoustic correlates, characterized by the quantitative or qualitative values in its phonological re...
متن کاملMultiple Reduced Phoneme Sets for Second Language Speech Recognition
This paper describes a novel method to improve the performance of second language speech recognition when the mother tongue of users is known. Considering that second language speech usually includes less fluent pronunciation and more frequent pronunciation mistakes, I propose using a reduced phoneme set generated by a phonetic decision tree (PDT)-based top-down sequential splitting method inst...
متن کاملUnsupervised Phoneme Segmentation Using Transformed Cepstrum Features
One of the basic problems in speech engineering is phoneme segmentation, that is, to divide a speech stream into a string of phonemes. Automatic Speech Recognition (ASR) models often require reliable phoneme segmentation in the initial training phase, and Text-to-Speech (TTS) systems need a large speech database with correct phoneme segmentation information for improving the performance. Human ...
متن کامل