Automatic Detection of Hyperarticulated Speech

Eugenio Ribeiro, Fernando Batista, Isabel Trancoso, Ricardo Ribeiro, David de Matos

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

Abstract

Hyperarticulation is a speech adaptation that consists of adopting a clearer form of speech in an attempt to improve recognition levels. However, it has the opposite effect when talking to ASR systems, as they are not trained with such kind of speech. We present approaches for automatic detection of hyperarticulation, which can be used to improve the performance of spoken dialog systems. We performed experiments on Let’s Go data, using multiple feature sets and two classification approaches. Many relevant features are speaker dependent. Thus, we used the first turn in each dialog as the reference for the speaker, since it is typically not hyperarticulated. Our best results were above 80 % accuracy, which represents an improvement of at least 11.6 % points over previously obtained results on similar data. We also assessed the classifiers’ performance in scenarios where hyperarticulation is rare, achieving around 98 % accuracy using different confidence thresholds.
Original languageEnglish
Title of host publicationAdvances in Speech and Language Technologies for Iberian Languages
Subtitle of host publicationThird International Conference, IberSPEECH 2016, Lisbon, Portugal, November 23-25, 2016, Proceedings
EditorsAlberto Abad, Alfonso Ortega, António Teixeira, Carmen García Mateo, Carlos D. Martínez Hinarejos, Fernando Perdigão, Fernando Batista, Nuno Mamede
PublisherSpringer Nature Switzerland AG
Pages182-191
ISBN (Electronic)978-3-319-49169-1
ISBN (Print)978-3-319-49168-4
DOIs
Publication statusPublished - Nov 2016
Externally publishedYes
EventInternational Conference on Advances in Speech and Language Technologies for Iberian Languages - Lisbon, Portugal
Duration: 23 Nov 201625 Nov 2016
https://link.springer.com/conference/iberspeech

Publication series

SeriesLecture Notes in Computer Science book series
Volumevolume 10077
SeriesLecture Notes in Artificial Intelligence (subseries)
Volumevolume 10077

Conference

ConferenceInternational Conference on Advances in Speech and Language Technologies for Iberian Languages
Abbreviated titleIberSPEECH 2016
CountryPortugal
CityLisbon
Period23/11/1625/11/16
Internet address

Keywords

  • Hyperarticulation
  • Speech
  • Let’s Go

Fingerprint Dive into the research topics of 'Automatic Detection of Hyperarticulated Speech'. Together they form a unique fingerprint.

Cite this