CNN-based phoneme classifier from vocal tract MRI learns embedding consistent with articulatory topology

K. G. Van Leeuwen, P. Bos, S. Trebeschi, M. J.A. Van Alphen, L. Voskuilen, L. E. Smeele, F. Van Der Heijden, R. J.J.H. Van Son

Research output: Contribution to conferencePaper

12 Downloads (Pure)

Abstract

Recent advances in real-time magnetic resonance imaging (rtMRI) of the vocal tract provides opportunities for studying human speech. This modality together with acquired speech may enable the mapping of articulatory configurations to acoustic features. In this study, we take the first step by training a deep learning model to classify 27 different phonemes from midsagittal MR images of the vocal tract. An American English database was used to train a convolutional neural network for classifying vowels (13 classes), consonants (14 classes) and all phonemes (27 classes) of 17 subjects. Classification top-1 accuracy of the test set for all phonemes was 57%. Error analysis showed voiced and unvoiced sounds often being confused. Moreover, we performed principal component analysis on the network's embedding and observed topological similarities between the network learned representation and the vowel diagram. Saliency maps gave insight into the anatomical regions most important for classification and show congruence with known regions of articulatory importance. We demonstrate the feasibility for deep learning to distinguish between phonemes from MRI. Network analysis can be used to improve understanding of normal articulation and speech and, in the future, impaired speech. This study brings us a step closer to the articulatory-to-acoustic mapping from rtMRI.

Original languageEnglish
Pages909-913
Number of pages5
DOIs
Publication statusPublished - 1 Jan 2019
Event20th Annual Conference of the International Speech Communication Association: Crossroads of Speech and Language, INTERSPEECH 2019 - Messecongress Graz, Graz, Austria
Duration: 15 Sep 201919 Sep 2019
Conference number: 20
https://www.interspeech2019.org/

Conference

Conference20th Annual Conference of the International Speech Communication Association: Crossroads of Speech and Language, INTERSPEECH 2019
Abbreviated titleINTERSPEECH 2019
CountryAustria
CityGraz
Period15/09/1919/09/19
Internet address

Keywords

  • Articulatory-to-acoustic mapping
  • Convolutional neural networks (CNN)
  • Deep learning
  • Magnetic resonance imaging (MRI)
  • Speech analysis
  • Vocal tract

Fingerprint Dive into the research topics of 'CNN-based phoneme classifier from vocal tract MRI learns embedding consistent with articulatory topology'. Together they form a unique fingerprint.

  • Cite this

    Van Leeuwen, K. G., Bos, P., Trebeschi, S., Van Alphen, M. J. A., Voskuilen, L., Smeele, L. E., ... Van Son, R. J. J. H. (2019). CNN-based phoneme classifier from vocal tract MRI learns embedding consistent with articulatory topology. 909-913. Paper presented at 20th Annual Conference of the International Speech Communication Association: Crossroads of Speech and Language, INTERSPEECH 2019, Graz, Austria. https://doi.org/10.21437/Interspeech.2019-1173