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

Odette Scharenborg, Danny Merkx

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

43 Downloads (Pure)

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.
Original languageEnglish
Title of host publicationProceedings of the Machine Learning in Speech and Language Processing Workshop
Place of PublicationHyderabad, India
Pages1-3
Number of pages3
Publication statusPublished - 2018
EventMachine Learning in Speech and Language Processing Workshop - Google offices, Hyderabad, India
Duration: 7 Sep 20187 Sep 2018

Workshop

WorkshopMachine Learning in Speech and Language Processing Workshop
Country/TerritoryIndia
CityHyderabad
Period7/09/187/09/18

Keywords

  • Convolutional Neural Network
  • spoken-word recognition
  • computational modelling
  • articulatory features

Fingerprint

Dive into the research topics of 'The Role of Articulatory Feature Representation Quality in a Computational Model of Human Spoken-Word Recognition'. Together they form a unique fingerprint.

Cite this