Abstract

We investigate the use of phonetic motor invariants (MIs), that is, recurring kinematic patterns of the human phonetic articulators, to improve automatic phoneme discrimination. Using a multi-subject database of synchronized speech and lips/tongue trajectories, we first identify MIs commonly associated with bilabial and dental consonants, and use them to simultaneously segment speech and motor signals. We then build a simple neural network-based regression schema (called Audio-Motor Map, AMM) mapping audio features of these segments to the corresponding MIs. Extensive experimental results show that (a) a small set of features extracted from the MIs, as originally gathered from articulatory sensors, are dramatically more effective than a large, state-of-the-art set of audio features, in automatically discriminating bilabials from dentals; (b) the same features, extracted from AMM-reconstructed MIs, are as effective as or better than the audio features, when testing across speakers and coarticulating phonemes; and dramatically better as noise is added to the speech signal. These results seem to support some of the claims of the motor theory of speech perception and add experimental evidence of the actual usefulness of MIs in the more general framework of automated speech recognition.

DOI

10.1371/journal.pone.0024055

Publication Date

2011-09-01

Publication Title

PLOS ONE

Volume

6

Publisher

PUBLIC LIBRARY SCIENCE

ISSN

1932-6203

Embargo Period

2024-11-22

Comments

affiliation: Castellini, C (Reprint Author), Univ Genoa, LIRA Lab, Genoa, Italy. Castellini, Claudio; Metta, Giorgio; Tavella, Michele, Univ Genoa, LIRA Lab, Genoa, Italy. Badino, Leonardo; Metta, Giorgio; Sandini, Giulio; Fadiga, Luciano, Italian Inst Technol, Genoa, Italy. Grimaldi, Mirko, Salento Univ, CRIL, Lecce, Italy. Fadiga, Luciano, Univ Ferrara, DSBTA, I-44100 Ferrara, Italy. article-number: e24055 keywords-plus: SPEECH-PERCEPTION; RECOGNITION research-areas: Science & Technology - Other Topics web-of-science-categories: Multidisciplinary Sciences author-email: claudio.castellini@dlr.de funding-acknowledgement: European Commission [NEST-5010, FP7-IST-250026] funding-text: The authors acknowledge the support of the European Commission project CONTACT (grant agreement NEST-5010) and SIEMPRE (grant agreement number FP7-IST-250026). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. number-of-cited-references: 31 times-cited: 0 journal-iso: PLoS One doc-delivery-number: 817OO unique-id: ISI:000294683900024

Additional Files
UoP_Deposit_Agreement v1.1 20160217.pdf (123 kB)
Download

Share

COinS