Perception of distributed coarticulatory properties of English /l/ and /r/

Abstract This paper investigates the salience of long-range coarticulatory effects in speech perception. English /l/ and /r/ have secondary articulations, resulting in resonances, which have been claimed to extend over several syllables and to have a domain larger than the phonological foot. For example, “a belly” differs from “a berry” not only in the liquid but also in the qualities of all three vowels. A progressive replacement experiment was conducted to test whether this type of distributed coarticulatory information is available to listeners in speech perception. It was found that subjects could correctly identify whether a word contained /l/ or /r/ even when the approximant and some surrounding vowels and consonants are replaced by noise. The experiment supports observations of the extent of secondary articulations, and provides reliable evidence that long-domain coarticulatory information about the /l/-/r/ distinction is perceptually available to listeners. While the specific phonetic nature of the long-range effects in this study is not clear, there are some candidate phenomena (such as F 2 of the surrounding context) that merit further study.

[1]  Osamu Fujimura,et al.  Allophonic variation in English /l/ and its implications for phonetic implementation , 1993 .

[2]  H. Magen.  Vowel‐to‐vowel coarticulation in English and Japanese , 1984 .

[3]  C A Fowler,et al.  Production and perception of coarticulation among stressed and unstressed vowels. , 1981, Journal of speech and hearing research.

[4]  P. Delattre,et al.  Comparing the Phonetic Features of English, French, German and Spanish , 1967 .

[5]  Shrikanth S. Narayanan,et al.  Toward articulatory-acoustic models for liquid approximants based on MRI and EPG data. Part I. The laterals , 1997 .

[6]  R. Meddis.  Statistics Using Ranks: A Unified Approach , 1984 .

[7]  Dennis Butler Fry.  The Physics of Speech , 1979 .

[8]  S. E. G. Öhman.  Perception of Segments of VCCV Utterances , 1966 .

[9]  Daniel Recasens,et al.  Long range coarticulation effects for tongue dorsum contact in VCVCV sequences , 1989, Speech Commun..

[10]  J. R. Firth,et al.  SOUNDS AND PEOSODIES , 1948 .

[11]  L. Raphael,et al.  Cross-sectional tongue shape and linguopalatal contact patterns in [s], [∫], [f], and [1] , 1992 .

[12]  Norman J. Lass,et al.  Principles of Experimental Phonetics , 1996 .

[13]  Terrance M. Nearey,et al.  The segment as a unit of speech perception , 1990 .

[14]  R. J. Lickley,et al.  Proceedings of the International Conference on Spoken Language Processing. , 1992 .

[15]  D. Whalen.  Coarticulation is largely planned , 1990 .

[16]  Adamantios I. Gafos,et al.  The Articulatory Basis of Locality in Phonology , 1999 .

[17]  Daniel Jones.  An outline of English phonetics , 1956 .

[18]  B. Repp.  On levels of description in speech research. , 1981, The Journal of the Acoustical Society of America.

[19]  D. Abercrombie,et al.  Elements of General Phonetics , 1967 .

[20]  H. Kucera,et al.  Computational analysis of present-day American English , 1967 .

[21]  R. M. Warren,et al.  Speech perception and phonemic restorations , 1971 .

[22]  S. Öhman.  Coarticulation in VCV Utterances: Spectrographic Measurements , 1966 .

[23]  John Local,et al.  Some Rhythm, Resonance, and Quality Variations in Urban Tyneside Speech. , 1986 .

[24]  Sarah Hawkins,et al.  Spread of CV and v-to-v coarticulation in british English: implications for the intelligibility of synthetic speech , 1994, ICSLP.

[25]  H. Magen.  The extent of vowel-to-vowel coarticulation in English and Japanese , 1997 .

[26]  J. B. Pickering,et al.  Vowel Perception and Production , 1994 .

[27]  Ronald W. Schafer,et al.  Digital Processing of Speech Signals , 1978 .

[28]  S. Manuel,et al.  The role of contrast in limiting vowel-to-vowel coarticulation in different languages. , 1990, The Journal of the Acoustical Society of America.