Standard

The Role of Articulatory Feature Representation Quality in a Computational Model of Human Spoken-Word Recognition. / Scharenborg, Odette; Merkx, Danny.

Proceedings of the Machine Learning in Speech and Language Processing Workshop. Hyderabad, India, 2018. p. 1-3.

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Harvard

Scharenborg, O & Merkx, D 2018, The Role of Articulatory Feature Representation Quality in a Computational Model of Human Spoken-Word Recognition. in Proceedings of the Machine Learning in Speech and Language Processing Workshop. Hyderabad, India, pp. 1-3, Machine Learning in Speech and Language Processing Workshop, Hyderabad, India, 7/09/18.

APA

Scharenborg, O., & Merkx, D. (2018). The Role of Articulatory Feature Representation Quality in a Computational Model of Human Spoken-Word Recognition. In Proceedings of the Machine Learning in Speech and Language Processing Workshop (pp. 1-3). Hyderabad, India.

Vancouver

Scharenborg O, Merkx D. The Role of Articulatory Feature Representation Quality in a Computational Model of Human Spoken-Word Recognition. In Proceedings of the Machine Learning in Speech and Language Processing Workshop. Hyderabad, India. 2018. p. 1-3

Author

Scharenborg, Odette ; Merkx, Danny. / The Role of Articulatory Feature Representation Quality in a Computational Model of Human Spoken-Word Recognition. Proceedings of the Machine Learning in Speech and Language Processing Workshop. Hyderabad, India, 2018. pp. 1-3

BibTeX

@inproceedings{e734e1ed8e274779a89c6a4bd797aa63,
title = "The Role of Articulatory Feature Representation Quality in a Computational Model of Human Spoken-Word Recognition",
abstract = "Fine-Tracker is a speech-based model of human speech recognition. While previous work has shown that Fine-Tracker is successful at modelling aspects of human spoken-word recognition, its speech recognition performance is not comparable to that of human performance, possibly due to suboptimal intermediate articulatory feature (AF) representations. This study investigates the effect of improved AF representations, obtained using a state-of-the-art deep convolutional network, on Fine-Tracker’s simulation and recognition performance: Although the improved AF quality resulted in improved speech recognition; it, surprisingly, did not lead to an improvement in Fine-Tracker’s simulation power.",
keywords = "Convolutional Neural Network, spoken-word recognition, computational modelling, articulatory features",
author = "Odette Scharenborg and Danny Merkx",
year = "2018",
language = "English",
pages = "1--3",
booktitle = "Proceedings of the Machine Learning in Speech and Language Processing Workshop",

}

RIS

TY - GEN

T1 - The Role of Articulatory Feature Representation Quality in a Computational Model of Human Spoken-Word Recognition

AU - Scharenborg, Odette

AU - Merkx, Danny

PY - 2018

Y1 - 2018

N2 - Fine-Tracker is a speech-based model of human speech recognition. While previous work has shown that Fine-Tracker is successful at modelling aspects of human spoken-word recognition, its speech recognition performance is not comparable to that of human performance, possibly due to suboptimal intermediate articulatory feature (AF) representations. This study investigates the effect of improved AF representations, obtained using a state-of-the-art deep convolutional network, on Fine-Tracker’s simulation and recognition performance: Although the improved AF quality resulted in improved speech recognition; it, surprisingly, did not lead to an improvement in Fine-Tracker’s simulation power.

AB - Fine-Tracker is a speech-based model of human speech recognition. While previous work has shown that Fine-Tracker is successful at modelling aspects of human spoken-word recognition, its speech recognition performance is not comparable to that of human performance, possibly due to suboptimal intermediate articulatory feature (AF) representations. This study investigates the effect of improved AF representations, obtained using a state-of-the-art deep convolutional network, on Fine-Tracker’s simulation and recognition performance: Although the improved AF quality resulted in improved speech recognition; it, surprisingly, did not lead to an improvement in Fine-Tracker’s simulation power.

KW - Convolutional Neural Network

KW - spoken-word recognition

KW - computational modelling

KW - articulatory features

M3 - Conference contribution

SP - 1

EP - 3

BT - Proceedings of the Machine Learning in Speech and Language Processing Workshop

CY - Hyderabad, India

ER -

ID: 47478570