This study uses articulography, the measurement of the position of tongue and lips during speech, as a tool to quantitatively assess the differences between pronunciations of native and non-native (Dutch) speakers of English. In our study, we focus on two pairs of English sound contrasts: /s/-/ʃ/ and /t/-/θ/. Our analysis focuses on the anterior-posterior position of the tongue tip during the p...

This paper is an analysis of lexical choices, pronunciation errors, and discourse features found in a corpus of student presentation speech. The speakers were Swedish natives studying Technical English. Particular emphasis is given to the pronunciation of the words most often used in academic texts. 93% of words used in the corpus came from the most frequent 2570 lexemes of academic written Eng...

This paper proposes a paradigm where commonly made segmental pronunciation errors are modeled as pair-wise confusions between two or more phonemes in the language that is being learnt. The method uses an ensemble of support vector machine classifiers with time varying Mel frequency cepstral features to distinguish between several pairs of phonemes. These classifiers are then applied to classify...

This paper describes research on automatic assessment of the pronunciation quality of spontaneous non-native adult speech. Since the speaking content is not known prior to the assessment, a two-stage method is developed to first recognize the speaking content based on non-native speech acoustic properties and then forced-align the recognition results with a reference acoustic model reflecting n...

This paper gives a detailed information on the design of the speech corpus for the purpose of developing an ASR-based pronunciation tutoring system. In the first place, assumptions on the structure of the corpus are presented. Then collection of text material, recordings and procedure of annotation of the resulting speech corpus are described. In the end, preliminary results of the analysis of ...

We have developed an original CALL system designed to detect and diagnose English pronunciation errors in Japanese learners’ speech. We have begun using the system in an English class at Kyoto University in Kyoto, Japan. The class consists of approximately 40 students, and the students have practiced pronunciation using this system on four separate occasions. Sixty minutes was allocated for eac...

In this paper the application of automatic speech recognition (ASR) technology in CAPT (Computer Aided Pronunciation Training) is addressed. A method to automatically generate the competitive lexicon, required by an ASR engine to compare the pronunciation of a target word with its correct and wrong phonetic realization, is presented. In order to enable the efficient deployment of CAPT applicati...

This study aims to determine the forms of errors in Arabic pronunciation at An-Nur Tangkit Islamic Boarding School, Sungai Gelam Muaro Jambi phonological level, factors pronunciation. The type research used is qualitative research. Collecting data from this using methods observation, interviews, and documentation. results indicate that occur Pondok Pesantren are grouped into 2: 1. Phoneme chang...

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...

This work reports on the construction of a rated database of spontaneous speech produced by second language (L2) learners of English. Spontaneous speech was collected from 28 L2 speakers representing six language backgrounds and five different proficiency levels. Speech was elicited using formats similar to that of the TOEFL iBT and the Speaking Proficiency English Assessment Kit (SPEAK) test. ...