Speech Contrast Research Articles

Acoustic Analysis of the Voiced-Voiceless Distinction in Dutch Tracheoesophageal Speech

Confusions between voiced and voiceless plosives and voiced and voiceless fricatives are common in Dutch tracheoesophageal (TE) speech. This study investigates (a) which acoustic measures are found to convey a correct voicing contrast in TE speech and (b) whether different measures are found in TE speech than in normal laryngeal (NL) speech. The main focus was on plosives. Nine TE and 5 NL speakers were included in this study. Fourteen acoustic measures were selected and analyzed. Comparisons were made between voiced and voiceless for the groups separately and between TE and NL speakers. Conditional inference trees were used to establish the most important distinguishing cue. TE speakers do not differ significantly from NL speakers, except for pitch-related measures. For plosives, all measures distinguished between voiced and voiceless for both speaker groups. The main distinguishing measure for the plosives was relative phonation time in the closure. TE and NL speakers differ less in the way the voiced-voiceless distinction is conveyed than expected. Further research is needed to show whether the main acoustic cue is also perceptually most relevant.

Rise time and formant transition duration in the discrimination of speech sounds: the Ba-Wa distinction in developmental dyslexia

Across languages, children with developmental dyslexia have a specific difficulty with the neural representation of the sound structure (phonological structure) of speech. One likely cause of their difficulties with phonology is a perceptual difficulty in auditory temporal processing (Tallal, 1980). Tallal (1980) proposed that basic auditory processing of brief, rapidly successive acoustic changes is compromised in dyslexia, thereby affecting phonetic discrimination (e.g. discriminating /b/ from /d/) via impaired discrimination of formant transitions (rapid acoustic changes in frequency and intensity). However, an alternative auditory temporal hypothesis is that the basic auditory processing of the slower amplitude modulation cues in speech is compromised (Goswami et al., 2002). Here, we contrast children's perception of a synthetic speech contrast (ba/wa) when it is based on the speed of the rate of change of frequency information (formant transition duration) versus the speed of the rate of change of amplitude modulation (rise time). We show that children with dyslexia have excellent phonetic discrimination based on formant transition duration, but poor phonetic discrimination based on envelope cues. The results explain why phonetic discrimination may be allophonic in developmental dyslexia (Serniclaes et al., 2004), and suggest new avenues for the remediation of developmental dyslexia.

The role of selective attention in the acquisition of English tense and lax vowels by native Spanish listeners: Comparison of three training methods.

This study investigates the role of two processes, cue enhancement (learning to attend to acoustic cues which characterize a speech contrast for native listeners) and cue inhibition (learning to ignore cues that do not), in the acquisition of the American English tense and lax ([i] and [ɪ]) vowels by native Spanish listeners. This contrast is acoustically distinguished by both vowel spectrum and duration. However, while native English listeners rely primarily on spectrum, inexperienced Spanish listeners tend to rely exclusively on duration. Twenty‐nine native Spanish listeners, initially reliant on vowel duration, received either enhancement training, inhibition training, or training with a natural cue distribution. Identification results demonstrated that, although listeners in all training groups increased their reliance on spectrum while decreasing reliance on duration, inhibitory and enhancement methods were significantly better than natural distribution training. Adaptive training was also more succe...

Clear speech production of Korean stops: Changing phonetic targets and enhancement strategies

The proposal that phonological contrast is enhanced through greater approximation of phonetic targets was investigated by comparing clear speech to conversational and citation-form speech produced by Korean speakers. The stop system of Korean is undergoing a sound change in which younger speakers produce the aspirated-lenis stop contrast differently from previous generations. Older speakers differentiate this contrast primarily with the acoustic correlate of voice onset time (VOT) and secondarily with F0. Younger speakers are merging VOT values for this contrast. As a result, the primary acoustic correlate is now F0 for younger speakers. These production differences likely indicate that younger speakers have developed different phonetic targets for stop production. These different phonetic targets were predicted to result in different enhancement patterns in clear speech: Younger speakers were predicted to enhance F0 differences, whereas older speakers were predicted to enhance VOT differences in clear speech. Results indicated that the older group solely used VOT to enhance the contrast in clear speech, whereas the younger group primarily used F0 but also demonstrated small VOT enhancement. These results indicate that clear speech enhancement strategies reflect phonetic targets. Older and younger speakers have different F0 and VOT targets and these different targets conditioned different enhancement strategies.

Auditory mismatch negativity for speech sound contrasts is modulated by language context

Auditory mismatch negativity is known to reflect language experience. This study wants to clarify whether this effect is dependent on language context. We compared German subjects' magnetic mismatch negativity in response to a non-native speech contrast presented with and without the context of a native contrast. The presence of the native contrast abolished the response to the nonnative sound at the left hemisphere and reduced the right-sided response. A significant context effect set in 170 ms after stimulus onset. As a control, the native contrast was increased by the context. This suggests that the auditory system adapts to the language context and modulates speech processing at early stages of information processing. This may facilitate elevated concentration on acoustic details that are linguistically relevant.

Developmental effects of multiple looks in speech sound discrimination.

The change/no-change procedure (J. E. Sussman & A. E. Carney, 1989), which assesses speech discrimination, has been used under the assumption that the number of stimulus presentations does not influence performance. Motivated by the tenets of the multiple looks hypothesis (N. F. Viemeister & G. H. Wakefield, 1991), work by R. F. Holt and A. E. Carney (2005) called this assumption into question (at least for adults): Nonsense syllable discrimination improved with more stimulus presentations. This study investigates the nature of developmental differences and the effects of multiple stimulus presentations in the change/no-change procedure. Thirty normal-hearing children, ages 4.0-5.9 years, were tested on 3 consonant-vowel contrasts at various signal-to-noise ratios using combinations of 2 and 4 standard and comparison stimulus repetitions. Although performance fell below that which is predicted by the multiple looks hypothesis in most conditions, discrimination was enhanced with more stimulus repetitions for 1 speech contrast. The relative influence of standard and comparison stimulus repetitions varied across the speech contrasts in a manner different from that of adults. Despite providing no additional sensory information, multiple stimulus repetitions enhanced children's discrimination of 1 set of nonsense syllables. The results have implications for models of developmental speech perception and assessing speech discrimination in children.

Principal Component Analyses and Scalp Distribution of the Auditory P150–250 and N250–550 to Speech Contrasts in Mexican and American Infants

We report a Principal Component Analysis (PCA) and the scalp distribution of the normalized peak amplitude values for speech-related auditory Event-related Potentials (ERP) P150–250 and N250–550 in 7-, 11-, and 20-month-old American infants learning English and in 10–13-month-old Mexican infants learning Spanish. After assessing the infant auditory ERP P-N complex using PCA, we evaluated the topographic distribution of each of the discriminatory phases to native and non-native CV-syllabic contrasts used in Spanish and English. We found that the first two Principal Components for each contrast type across ages showing a maximization of differences between the P150–250 and the N250–550 waves, explain more than 70% of the variance. The scalp distributions of the P150–250 and N250–550 components also differed, the P150–250 showing a frontal and anterior temporal distribution, and the N250–550 a more posterior distribution. The older infants showed a broader distribution of responses, particularly for the N250–550. There were no differences in the topographies of the components between same-aged Mexican and American infants. We discuss the perceptual/linguistic functions that each component may reflect during development and across the two cultures.

Training Chinese dialect listeners to perceive English /n/ and /l/

Dialects and Mandarin subdialects in certain areas of southern China do not distinguish between /n/ and /l/. Native speakers of these dialects usually have difficulty distinguishing these two sounds when learning English, even after years of experience with the language. The goal of the present research is to develop a set of procedures that would be effective in training these speakers to learn this non-native speech contrast in English. Forty listeners were recruited from 3 dialect areas (6 Gan, 11 Jianghuai, and 23 Xinan). Listeners were instructed to identify the initial consonant in an English word spoken by 3 native speakers of American English. A pre- and a post-test design were used to assess the effect of a 4-day training. A generalization test containing the same stimuli spoken by 2 new talkers was also carried out on the same day of the post-test. Overall, subjects’ performance improved significantly from the pre-test to the post-test and there was no significant group difference. Additionally, there was no significant difference between their performance of /n/ and /l/ either before or after training. Vowel contexts and syllable structures are also found to have affected listeners’ perception in the post-test.

The effect of a hearing aid noise reduction algorithm on the acquisition of novel speech contrasts

Audiologists are reluctant to prescribe digital hearing aids with active digital noise reduction (DNR) to pre-verbal children due to their potential for an adverse effect on the acquisition of language. The present study investigated the relation between DNR and language acquisition by modeling pre-verbal language acquisition using adult listeners presented with a non-native speech contrast. Two groups of normal-hearing, monolingual Anglophone subjects were trained over four testing sessions to discriminate novel, difficult to discriminate, non-native Hindi speech contrasts in continuous noise, where one group listened to both speech items and noise processed with DNR, and where the other group listened to unprocessed speech in noise. Results did not reveal a significant difference in performance between groups across testing sessions. A significant learning effect was noted for both groups between the first and second testing sessions only. Overall, DNR does not appear to enhance or impair the acquisition of novel speech contrasts by adult listeners.

The preservation of schwa in the converging phonological system of Frenchville (PA) French

The phonological system of the French of Frenchville, Pennsylvania (USA) demonstrates a dramatic case of transfer in the latest (and last) generation of bilingual French–English speakers: the mid front round vowels, [œ] and [ø], have often been replaced by the English rhoticized schwa as found in the word sir. However, French schwa, which is arguably phonetically non-distinct from the mid front round vowels, does not participate fully in this merger. This result is unexpected given both the phonetic identity of schwa and [ø], and the fact that our speakers are not literate in French and, as such, have no access to the differential orthographic representations manifest between schwa and the mid front round vowels. The data argue strongly that schwa is, in some sense, “real” for these speakers. Based on a phonetic analysis of the vowels under consideration, we argue that transfer between two sound systems cannot be perceived as a simple case of phonetic replacement. Instead, transfer or convergence with English must be viewed as a systemic process that preserves contrast in unexpected ways. In the case at hand, the data suggest that the traditional separation between the phonetic and phonological levels of grammar cannot be maintained as each level contributes to both provoking merger and maintaining contrast in bilingual speech.

Child codeswitching and adult content contrasts

This paper discusses links between the conditions of developing child bilingualism and the adult outcomes in semiotic contrast in elicited speech and codeswitching. Analysis of interaction of children raised as bilinguals shows that from the beginning they can recognize the appropriate language for addressees. When the lexical repertoire is inadequate, borrowing occurs, and codeswitching of longer segments appears before age two. Throughout childhood and adolescence, codeswitching has increasing interactional functions as children's pragmatic skills grow. Some adult codeswitching relies on semiotic differences implied by language. It is likely that both such codeswitching and the dual selves shown in elicited discourse in different languages are limited to specific sociolinguistic situations and personal histories. These include education in a second language, adult immigration, and frequent contact with a monolingual community.

The mismatch negativity (MMN) response to complex tones and spoken words in individuals with aphasia

Background: The mismatch negativity (MMN) is a fronto-centrally distributed event-related potential (ERP) that is elicited by any discriminable auditory change. It is an ideal neurophysiological tool for measuring the auditory processing skills of individuals with aphasia because it can be elicited even in the absence of attention. Previous MMN studies have shown that acoustic processing of tone or pitch deviance is relatively preserved in aphasia, whereas the basic acoustic processing of speech stimuli can be impaired (e.g., auditory discrimination). However, no MMN study has yet investigated the higher levels of auditory processing, such as language-specific phonological and/or lexical processing, in individuals with aphasia.Aims: The aim of the current study was to investigate the MMN responses of normal and language-disordered subjects to tone stimuli and speech stimuli that incorporate the basic auditory processing (acoustic, acoustic-phonetic) levels of non-speech and speech sound processing, and also the language-specific phonological and lexical levels of spoken word processing. Furthermore, this study aimed to correlate the aphasic MMN data with language performance on a variety of tasks specifically targeted at the different levels of spoken word processing.Methods & Procedures: Six adults with aphasia (71.7 years ±3.0) and six healthy age-, gender-, and education-matched controls (72.2 years ±5.4) participated in the study. All subjects were right-handed and native speakers of English. Each subject was presented with complex harmonic tone stimuli, differing in pitch or duration, and consonant-vowel (CV) speech stimuli (non-word /de:/ versus real word /deI/). The probability of the deviant for each tone or speech contrast was 10%. The subjects were also presented with the same stimuli in behavioural discrimination tasks, and were administered a language assessment battery to measure their auditory comprehension skills.Outcomes & Results: The aphasic subjects demonstrated attenuated MMN responses to complex tone duration deviance and to speech stimuli (words and non-words), and their responses to the frequency, duration, and real word deviant stimuli were found to strongly correlate with performance on the auditory comprehension section of the Western Aphasia Battery (WAB). Furthermore, deficits in attentional lexical decision skills demonstrated by the aphasic subjects correlated with a word-related enhancement demonstrated during the automatic MMN paradigm, providing evidence to support the “word advantage effect”, thought to reflect the activation of language-specific memory traces in the brain for words.Conclusions: These results indicate that the MMN may be used as a technique for investigating general and more specific auditory comprehension skills of individuals with aphasia, using speech and/or non-speech stimuli, independent of the individual's attention. The combined use of the objective MMN technique and current clinical language assessments may result in improved rehabilitative management of aphasic individuals.

Perception of a non-native speech contrast: Voiced and voiceless stops as perceived by Tamil speakers

The effect of linguistic experience plays a significant role in how speech sounds are perceived. The findings of many studies imply that the perception of non-native contrasts depends on their status in the native language of the listener. Tamil is a language with a single voicing category. All stop consonants in Tamil are phonemically voiceless, though allophonic voicing has been observed in spoken Tamil. The present study examined how native Tamil speakers and English controls perceived voiced and voiceless bilabial, alveolar, and velar stops in English. Voice onset time (VOT) was manipulated for editing of naturally produced stimuli with increasingly longer continuum. Perceptual data was collected from 16 Tamil and 16 English speakers. Experiment 1 was an AX task in which subjects responded same or different to 162 pairs of stimuli. Experiment 2 was a forced choice ID task in which subjects identified 99 individually presented stimuli as pa, ta, ka or ba, da, ga. Experiments show statistically significant differences between Tamil and English speakers in their perception of English stop consonants. Results of the study imply that the allophonic status of voiced stops in Tamil does not aid the Tamil speakers in perceiving phonemically voiced stops in English.

Nonsensory factors in speech perception

The nature of developmental differences was examined in a speech discrimination task, the change/no-change procedure, in which a varying number of speech stimuli are presented during a trial. Standard stimuli are followed by comparison stimuli that are identical to or acoustically different from the standard. Fourteen adults and 30 4- and 5-year-old children were tested with three speech contrast pairs at a variety of signal-to-noise ratios using various numbers of standard and comparison stimulus presentations. Adult speech discrimination performance followed the predictions of the multiple looks hypothesis [N. F. Viemeister and G. H. Wakefield, J. Acoust. Soc. Am. 90, 858–865 (1991)]; there was an increase in d by a factor of 1.4 for a doubling in the number of standard and comparison stimulus presentations near d values of 1.0. For children, increasing the number of standard stimuli improved discrimination performance, whereas increasing the number of comparisons did not. The multiple looks hypothesis did not explain the children’s data. They are explained more parsimoniously by the developmental weighting shift [Nittrouer et al., J. Acoust. Soc. Am. 101, 2253–2266 (1993)], which proposes that children attend to different aspects of speech stimuli from adults. [Work supported by NIDCD and ASHF.]

Teaching the /r/–/l/ discrimination to Japanese adults: behavioral and neural aspects

Several studies have been conducted to address the learning of a nonnative speech contrast in adulthood, using native speakers of Japanese and the English /r/–/l/ contrast. Japanese adults were asked to identify contrasting /r/–/l/ stimuli (e.g., “rock–lock”). An adaptive training regime starting with initially easy stimuli was contrasted with a fixed training regime using difficult stimuli, with some subjects receiving feedback on the correctness of their responses and others receiving no feedback in both conditions. After three and six sessions of training, subjects received tests assessing identification and discrimination of /r/–/l/ stimuli as well as generalization. In all cases except fixed training without feedback, subjects showed clear evidence of learning, and several indicators suggested that training affects speech perception, rather than simply auditory processes. Neuroimaging studies currently underway are examining the neural basis of these findings.

Success and failure in teaching the [r]-[l] contrast to Japanese adults: tests of a Hebbian model of plasticity and stabilization in spoken language perception.

A Hebbian model of learning predicts that adults may be able to acquire a nonnative speech contrast if they are trained with stimuli that are exaggerated to make them perceptually distinct. To test these ideas, we asked Japanese adults to identify contrasting [r]-[l] stimuli (e.g., rock-lock) in two training conditions. In the adaptive condition, the [r]-[l] contrast was exaggerated at first and then adjusted to maintain accurate identification. In the fixed condition, a fixed pair of stimuli were used that were distinguishable by native English speakers but difficult for the Japanese learners to discriminate. To examine whether feedback contributes to learning, we ran separate groups with and without feedback in the fixed and the adaptive conditions. Without feedback, 3 days of adaptive training produced substantial improvements, but 3 days of fixed training produced no benefit relative to control, consistent with the Hebbian account. With feedback, both fixed and adaptive training led to robust improvements, and the benefit of training transferred to a second continuum (e.g., road-load). The results are consistent with Hebbian models that are augmented to be sensitive to feedback.

Mismatch negativity shows that 3–6-year-old children can learn to discriminate non-native speech sounds within two months

Using 3–6-year-old children as subjects, we describe the neural plasticity accompanying the concurrent learning of a foreign language in a natural environment. Children were monitored for 6 months as they either enrolled in schools or daycare centers where only Finnish was spoken (Control group) or as they joined a French school or a daycare center where French was spoken 50–90% of the time (Experimental group). Whereas mismatch negativity (MMN)—a brain's electrical change-detection response—for a French speech contrast was initially absent or very small in both groups, it was conspicuous 2 months after Finnish children had joined a French kindergarten. Consequently, the data suggest that youngsters can learn to distinguish non-native speech sounds in natural language environment without any special training in just a couple of months. Accordingly, these data herald the vast potential MMN may entail for studying language learning, especially in situations where behavioral responses cannot be readily elicited.

Cross-language speech perception: Evidence for perceptual reorganization during the first year of life

Previous work in which we compared English infants, English adults, and Hindi adults on their ability to discriminate two pairs of Hindi (non-English) speech contrasts has indicated that infants discriminate speech sounds according to phonetic category without prior specific language experience (Werker, Gilbert, Humphrey, & Tees, 1981), whereas adults and children as young as age 4 (Werker & Tees, in press), may lose this ability as a function of age and or linguistic experience. The present work was designed to (a) determine the generalizability of such a decline by comparing adult English, adult Salish, and English infant subjects on their perception of a new non-English (Salish) speech contrast, and (b) delineate the time course of the developmental decline in this ability. The results of these experiments replicate our original findings by showing that infants can discriminate non-native speech contrasts without relevant experience, and that there is a decline in this ability during ontogeny. Furthermore, data from both cross-sectional and longitudinal studies shows that this decline occurs within the first year of life, and that it is a function of specific language experience.

The effect of spectral contrast enhancement on speech perception in hearing-impaired listeners

The effect of enhancing the spectral contrast of clean and noisy speech on intelligibility and listening comfort in hearing-impaired listeners is explored. Spectral contrast refers to the difference between the amplitude of spectral peaks and valleys in a short temporal window. A method of enhancing spectral contrast similar to that suggested by Bunnell [H. T. Bunnell, J. Acoust. Soc. Am. 88, 2546–2556 (1990)] was implemented in a simulated hearing aid with fast acting multiband wide-dynamic-range compression. The aid’s frequency response was adjusted to accommodate the individual listeners’ hearing losses. The effect of several degrees of spectral contrast enhancement on phoneme identification in quiet and in the presence of speech-spectrum-shaped noise was measured. In addition, listeners’ preferences for several degrees of spectral contrast enhancement were assessed in an A/B comparison of spectrally enhanced connected speech in quiet and in the presence of cafeteria noise. The experimental results will be reported and their implications for the use of spectral contrast enhancements in hearing aids will be discussed.

Reliability of the Mismatch Negativity in the Responses of Individual Listeners

The mismatch negativity (MMN) was recorded from 12 normal adults during four biweekly sessions. Responses were elicited by a synthetically generated speech contrast (/dalpha/-/galpha/) that all listeners discriminated with at least 90 percent accuracy. Standard and deviant waveforms were replicable across sessions for all listeners; however, replicability of the derived difference waveforms was poor. Of greater importance, the MMN identification rate was too low (29%) to allow reliability to be evaluated. The implications that these findings may have on clinical applicability are discussed.