ASR-Free Pronunciation Assessment

Cao, Shuai; Liu, Zhixin; Li, Lantian; Tang, Zhiyuan; Wang, Dong; Zheng, Thomas Fang

doi:10.21437/interspeech.2020-2623

Cited by 16 publications

(6 citation statements)

References 20 publications

(23 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…GoP was further extended with Deep Neural Networks (DNNs), replacing Hidden Markov Model (HMM) and Gaussian Mixture Model (GMM) techniques for acoustic modeling [4,5]. Cheng et al [8] improved the performance of GoP with the latent representation of speech extracted in an unsupervised way.…”

Section: Related Workmentioning

confidence: 99%

Mispronunciation Detection in Non-Native (L2) English with Uncertainty Modeling

Korzekwa

Lorenzo-Trueba

Zaporowski

et al. 2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

A common approach to the automatic detection of mispronunciation in language learning is to recognize the phonemes produced by a student and compare it to the expected pronunciation of a native speaker. This approach makes two simplifying assumptions: a) phonemes can be recognized from speech with high accuracy, b) there is a single correct way for a sentence to be pronounced. These assumptions do not always hold, which can result in a significant amount of false mispronunciation alarms. We propose a novel approach to overcome this problem based on two principles: a) taking into account uncertainty in the automatic phoneme recognition step, b) accounting for the fact that there may be multiple valid pronunciations. We evaluate the model on non-native (L2) English speech of German, Italian and Polish speakers, where it is shown to increase the precision of detecting mispronunciations by up to 18% (relative) compared to the common approach.

show abstract

Section: Related Workmentioning

confidence: 99%

Mispronunciation Detection in Non-Native (L2) English with Uncertainty Modeling

Korzekwa

Lorenzo-Trueba

Zaporowski

et al. 2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

show abstract

“…In addition, the pronunciation practice takes place in a "safe" PUPIL: International Journal of Teaching, Education and Learning ISSN 2457-0648 34 environment, where the learner is not exposed to the evaluation of their peers, and the feedback is individualized to their particular needs (Dai & Wu, 2021;Evers & Chen, 2022). Automatic speech recognition technology has been one of the critical components of pronunciation assessment (Cheng et al, 2020) since the 1990s. According to the authors, pronunciation assessment takes place in three steps: segmentation of speech into smaller unitspronunciation analysis of the speech unitscalculation of the score (comp.…”

Section: Literature Reviewmentioning

confidence: 99%

Automatic Pronunciation Assessment in a Flipped Classroom Context

Vancova

2023

pijtel

View full text Add to dashboard Cite

Automatic pronunciation assessment programs allow learners to practice pronunciation independently and receive immediate feedback based on acoustic analysis of their speech compared to a pronunciation model. The programs display a high level of reliability when assessing speakers’ pronunciation accuracy. The paper aims to present the results of an investigation into a recent experimental application of an automatic pronunciation assessment online program used within the flipped classroom of English phonetics and phonology course taught to foreign language pre-service teachers of English. Flipped learning approach was selected to offload the preliminary activities and tasks that do not require the immediate attention of the teacher to an online learning management system. Thus, course participants’ face-to-face learning could be enhanced and more effective. The participants could practice and receive tailored feedback from an online program before the face-to-face class. After the individual pronunciation practice, they could discuss more complex questions with the lecturer and, in collaboration with fellow students. The paper will present the survey results among the course participants who evaluated their experience with automatic pronunciation evaluation. Special attention is paid to the potential use of such programs for developing linguistic competence.

show abstract

“…One important reason for the lack of research in this area concerns the technical challenge of automatically recording and assessing speech to use in real-time adaptive learning systems. However, methods to automatically score pronunciation accuracy in real time currently exist (e.g., Moustroufas and Digalakis, 2007 ; Neri et al, 2008 ; Cheng et al, 2020 , and see www.emotech.ai ) and pilot data from our lab shows promising results for the application of such methods in adaptive, speech-based learning systems ( Wilschut et al, 2021 ). In the current study, we will further examine how to use speech in adaptive learning systems.…”

Section: Introductionmentioning

confidence: 96%

Benefits of Adaptive Learning Transfer From Typing-Based Learning to Speech-Based Learning

Wilschut

Sense

Velde

et al. 2021

Front. Artif. Intell.

View full text Add to dashboard Cite

Memorising vocabulary is an important aspect of formal foreign-language learning. Advances in cognitive psychology have led to the development of adaptive learning systems that make vocabulary learning more efficient. One way these computer-based systems optimize learning is by measuring learning performance in real time to create optimal repetition schedules for individual learners. While such adaptive learning systems have been successfully applied to word learning using keyboard-based input, they have thus far seen little application in word learning where spoken instead of typed input is used. Here we present a framework for speech-based word learning using an adaptive model that was developed for and tested with typing-based word learning. We show that typing- and speech-based learning result in similar behavioral patterns that can be used to reliably estimate individual memory processes. We extend earlier findings demonstrating that a response-time based adaptive learning approach outperforms an accuracy-based, Leitner flashcard approach in learning efficiency (demonstrated by higher average accuracy and lower response times after a learning session). In short, we show that adaptive learning benefits transfer from typing-based learning, to speech based learning. Our work provides a basis for the development of language learning applications that use real-time pronunciation assessment software to score the accuracy of the learner’s pronunciations. We discuss the implications for our approach for the development of educationally relevant, adaptive speech-based learning applications.

show abstract

ASR-Free Pronunciation Assessment

Cited by 16 publications

References 20 publications

Mispronunciation Detection in Non-Native (L2) English with Uncertainty Modeling

Mispronunciation Detection in Non-Native (L2) English with Uncertainty Modeling

Automatic Pronunciation Assessment in a Flipped Classroom Context

Benefits of Adaptive Learning Transfer From Typing-Based Learning to Speech-Based Learning

Contact Info

Product

Resources

About