Suprasegmental Information Research Articles

Constraints of lexical stress on lexical access in English: evidence from native and non-native listeners.

Four cross-modal priming experiments and two forced-choice identification experiments investigated the use of suprasegmental cues to stress in the recognition of spoken English words, by native (English-speaking) and non-native (Dutch) listeners. Previous results had indicated that suprasegmental information was exploited in lexical access by Dutch but not by English listeners For both listener groups, recognition of visually presented target words was faster, in comparison to a control condition, after stress-matching spoken primes, either monosyllabic (mus- from MUsic/muSEum) or bisyl-word recognition labic (admi-from ADmiral/admiRAtion). For native listeners, the effect of stress-mismatching bisyllabic primes was not different from that of control primes, but mismatching monosyllabic primes produced partial facilitation. For non-native listeners, both bisyllabic and monosyllabic stress-mismatching primes produced partial facilitation. Native English listeners thus can exploit suprasegmental information in spoken-word recognition, but information from two syllables is used more effectively than information from one syllable. Dutch listeners are less proficient at using suprasegmental information in English than in their native language, but, as in their native language, use mono- and bisyllabic information to an equal extent. In forced-choice identification. Dutch listeners outperformed native listeners at correctly assigning a monosyllabic fragment (e.g.. mus-) to one of two words differing in stress.

A Case Study of Acquired Dyslexia and Dysgraphia in Cantonese: Evidence for Nonsemantic Pathways for Reading and Writing Chinese

We report a Cantonese-speaking brain-damaged patient, CML, who demonstrates better oral reading than oral naming and better writing to dictation than written naming. Such dissociations are taken as evidence for nonsemantic routes for the production of spoken and written Chinese words. The occurrence of tonal errors in CML's reading aloud informs us about the structure of phonological representations. We propose that it is a nonlinear structure, similar to that which has been proposed in tonal phonology of Chinese. Specific impairment to these representations may lead to a dissociation between segmental and suprasegmental information. Finally, the similarity of CML's neologistic responses in oral and written naming, and her production of tonal errors in writing to dictation and written naming, favour the mediated version of the nonsemantic pathway for writing more than the unmediated pathway.

Voornaam is not (really) a homophone: Lexical prosody and lexical access in Dutch.

Four experiments examined Dutch listeners' use of suprasegmental information in spoken-word recognition. Isolated syllables excised from minimal stress pairs such as VOORnaam/voorNAAM could be reliably assigned to their source words. In lexical decision, no priming was observed from one member of minimal stress pairs to the other, suggesting that the pairs' segmental ambiguity was removed by suprasegmental information. Words embedded in nonsense strings were harder to detect if the nonsense string itself formed the beginning of a competing word, but a suprasegmental mismatch to the competing word significantly reduced this inhibition. The same nonsense strings facilitated recognition of the longer words of which they constituted the beginning, but again the facilitation was significantly reduced by suprasegmental mismatch. Together these results indicate that Dutch listeners effectively exploit suprasegmental cues in recognizing spoken words. Nonetheless, suprasegmental mismatch appears to be somewhat less effective in constraining activation than segmental mismatch.

Functional Heterogeneity of Inferior Frontal Gyrus Is Shaped by Linguistic Experience

A crosslinguistic, positron emission tomography (PET) study was conducted to determine the influence of linguistic experience on the perception of segmental (consonants and vowels) and suprasegmental (tones) information. Chinese and English subjects (10 per group) were presented binaurally with lists consisting of five Chinese monosyllabic morphemes (speech) or low-pass-filtered versions of the same stimuli (nonspeech). The first and last items were targeted for comparison; the time interval between target tones was filled with irrelevant distractor tones. A speeded-response, selective attention paradigm required subjects to make discrimination judgments of the target items while ignoring intervening distractor tones. PET scans were acquired for five tasks presented twice: one passive listening to pitch (nonspeech) and four active (speech = consonant, vowel, and tone; nonspeech = pitch). Significant regional changes in blood flow were identified from comparisons of group-averaged images of active tasks relative to passive listening. Chinese subjects show increased activity in left premotor cortex, pars opercularis, and pars triangularis across the four tasks. English subjects, on the other hand, show increased activity in left inferior frontal gyrus regions only in the vowel task and in right inferior frontal gyrus regions in the pitch task. Findings suggest that functional circuits engaged in speech perception depend on linguistic experience. All linguistic information signaled by prosodic cues engages left-hemisphere mechanisms. Storage and executive processes of working memory that are implicated in phonological processing are mediated in discrete regions of the left frontal lobe.

PHONOLOGICAL PRIMING IN CANTONESE SPOKEN-WORD PROCESSING

An auditory priming experiment with 28 native Cantonese speakers was conducted to examine the effects of phonological overlap between monosyllabic target words and preceding primes. In the experiment, listeners heard a pair of Cantonese monosyllables (prime-target) with 250 milliseconds inter-stimulus interval in between and they were instructed to name aloud the second item (target). Results indicated that facilitation was observed in two conditions: (1) shared rime and tone between primes and targets and (2) shared onset and rime between primes and targets. However, when shared the onset and tone between the monosyllabic prime and target, an inhibition effect was observed. These findings are discussed in terms of the segmental and supra-segmental information of Cantonese.

Emerging Integration of Sequential and Suprasegmental Information in Preverbal Speech Segmentation

5 studies examined contributions of syllable-ordering and rhythmic properties of syllable strings to 6- and 9-month-old infants' speech segmentation. A pair of methods measuring complementary properties of representational units was used: a noise detection task sensitive to perceived cohesiveness of pairs of syllables, and a discrimination maintenance task sensitive to compactness of representations of syllable pairs. For 9-month-olds, results show that a key pair of syllables was represented as a unit when the grouping of these syllables was supported by correlated regularities of ordering and rhythm in the set of stimulus strings, but not when such grouping was supported by only rhythmic or only syllable-ordering regularity. For 6-month-olds, results show that a key pair of syllables was represented as a unit whenever grouping was supported by rhythmic regularity in the stimulus strings, regardless of whether syllable-ordering regularity was also present. Thus, whereas 9-month-olds appear to be capable of integrating sequential and suprasegmental information in forming worldlike (multisyllabic) phonological percepts, 6-month-olds are not. The emergence of integrative abilities portends increased efficiency in speech processing and may contribute to the formation and use of an initial lexicon.

Emerging Integration of Sequential and Suprasegmental Information in Preverbal Speech Segmentation

Emerging Integration of Sequential and Suprasegmental Information in Preverbal Speech Segmentation

Choice of speech features for tactile presentation to the profoundly deaf

Measurements have been made, using acoustic presentation of stimuli, to compare a variety of speech-derived signals (amplitude envelope, voice fundamental frequency, second-formant frequency, zero-crossing frequency, and an "on/off" control derived from the amplitude envelope via a comparator) as to their suitability for tactile presentation to the profoundly deaf as an aid to speech reception. Segmental (phonemic) information was conveyed adequately by all five signals; suprasegmental (stress) information was conveyed very well by voice fundamental frequency, and significantly less well by the other signals. The best choice of speech features for presentation via a practical tactile aid is discussed.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

De Waarneming van Klemtoon bij NT2-Studenten

Recently several research groups that are concerned with language acquisition have claimed, that suprasegmental information contributes to the learnability of natural languages. Specifically, it is claimed, that suprasegmentals enable the language learner to perceive and distinguish linguistic patterns in the stream of speech (Morgan et al, 1987; Mehler et al, 1988; Hirsch-PAsek, 1987). It has been shown, that suprasegmentals are immediately accessable to babies and require no learning process. This finding is compatible with psychological research that has shown, that the sensoric system of babies structures perception immediately after birth (see Van Geert, 1983 for an overview). The present study deals with the use of suprasegmentals in adult second language learners. For several reasons it may be wondered whether adults have the same benefit from suprasegmentals like newborns. Therefore an experiment is conducted in order to investigate the ability of adult second language learners to distinguish several stress-patterns in Dutch. Several stress-patterns in two- and three syllabic words were investigated. The results indicate, that both groups of second language learners use a lexical and a perceptual strategy, that in the case of Level 1 students leads to a confusion of perceptual and lexical strategies, whereas Level Π students appear to seperate these strategies in a more appropriate way. The results are discussed in the perspective of language teaching. It is concluded, that the development of perceptual strategies need more attention in the classroom, since they might enable the student more effectively to explore a second language in communicative situations.

The role of talker-specific information in memory for spoken sentences

Contrary to most traditional theories of speech perception, talker-specific information is not filtered out or ‘‘lost’’ during perceptual analysis, but is retained in the processing and representation of spoken language. Effects of talker-specific information have been found for words and phonemes spoken in isolation [J. W. Mullennix and D. B. Pisoni, J. Acoust. Soc. Am. 85, 365–378 (1990); Goldinger etal., J. Exp. Psychol.: Learn., Mem. Cog. 17, 152–162 (1991)]. However, the processing of spoken sentences may be quite different from isolated words due to coarticulation across words, suprasegmental information, and syntactic and semantic constraints. The present experiment investigates whether the effects of talker-specific information on the processing of sentences are similar to those found for the processing of isolated words and phonemes. Subjects first transcribed lists of sentences in either a single-talker or multiple-talker presentation format; they were given a probed recall test to assess memory for the sentences. While transcription accuracy was higher in the single-talker than the multiple-talker condition, cued recall performance for correctly transcribed sentences was not significantly different between the two presentation conditions. Implications of these findings for current theories of spoken language processing will be discussed in the context of recent episodic and multiple-trace models of human memory.

Freedoms and constraints in computational prosody modeling from speech corpora

With the growth of the availability of large speech corpora, statistical models of prosody control have been studied intensively. These computational models have improved the naturalness of synthetic speech and are expected to provide additional supra-segmental information for speech recognition. Though only conventional statistical methods such as linear regression, regression trees, or neural nets have been employed in these computational models, their success has been due to efforts to accommodate observed prosodic characteristics and qualitatively known control mechanisms. In this talk, a review will be presented on how observed prosodic characteristics have been modeled with these statistical tools and also how new statistical models can be designed to cope with the insufficiencies of conventional models. It is expected that the investigation of well-constrained models and their constraints will lead one to more efficient computational models and deeper understanding of prosody control mechanism through these modeling procedures.

Auditory brainstem implant: II. Postsurgical issues and performance.

The auditory brainstem implant (ABI) restores some hearing sensations to patients deafened by bilateral acoustic tumors. Electrodes are stable for more than 10 years. In most cases nonauditory side effects can be avoided by judicious selection of the stimulating waveform and electrode configuration. Most perceptual measurements demonstrate that the ABI produces psychophysical and speech performance similar to that of single-channel cochlear implants. ABI patients receive suprasegmental information in speech and significant enhancement of speech understanding when the sound from the ABI is combined with lipreading.

Processing interactions between segmental and suprasegmental information in native speakers of English and Mandarin Chinese.

The processing interactions between segmental and suprasegmental information in native speakers of English and Mandarin Chinese were investigated in a speeded classification task. Since in Chinese, unlike in English, tones convey lexically meaningful information, native speakers of these languages may process combinations of segmental and suprasegmental information differently. Subjects heard consonant-vowel syllables varying on a consonantal (segmental) dimension and either a Mandarin Chinese or constant-pitch (non-Mandarin) suprasegmental dimension. The English listeners showed mutual integrality with the Mandarin Chinese stimuli, but not the constant-pitch stimuli. The native Chinese listeners processed these dimensions with mutual integrality for both the Mandarin Chinese and the constant-pitch stimuli. These results were interpreted in terms of the linguistic function and the structure of suprasegmental information in Chinese and English. The results suggest that the way listeners perceive speech depends on the interaction between the structure of the signal and the processing strategies of the listener.

Tone and intonation in Mandarin babytalk to presyllabic infants: Comparison with registers of adult conversation and foreign language instruction

ABSTRACTSix native speakers of Mandarin Chinese recorded 140 preselected utterances in three role-play contexts that differentially elicited registers of babytalk to presyllabic infants (BTP), foreign language instruction (FLI), and adult conversation (AC). Sound spectrograms were used to obtain 10 measures of fundamental frequency (Fo) patterns for comparisons among the three registers. In FLI, the speakers expanded Fo patterns in time and Fo range in comparison with AC. They clarified lexical tonal information and seemed to reduce suprasegmental information. In BTP, the speakers raised peak and minimum Fo, reduced the rate of Fo fluctuations, and increased the proportion of terminal rising contours. The speakers reduced, neglected, or modified lexical tonal information in favor of simplified and clarified intonation contours. The significance of the results is discussed in relation to tone acquisition in children and to a universal intuitive didactic competence in caretakers.

Influences of contextual and local temporal information on perception of a stop consonant after [s

Much of the perceptual information for a stop consonant in an [s]-stop cluster inheres in temporal properties of the acoustic signal local to the period of low energy corresponding to stop closure. However, stop percepts in this context are also influenced by the rate of speech in a precursor phrase to which the cluster is appended. The two experiments in this paper attempt to explore the dissociability of these local and longer-term perceptual influences, and to investigate the interactions between them. The results of the experiments reveal a complex pattern of interactions which suggests that a simple distinction between segmental and suprasegmental temporal information for phonetic contrasts is unrealistic.

Multichannel Cochlear Implant: The New York University/Bellevue Experience

A total of nine patients have been implanted at the New York University/Bellevue Medical Center with the Nucleus multichannel cochlear implant. The patients ranged in age from 21 to 62 years, with a mean age of 38.7 years. All were postlingually deafened with bilateral profound sensorineural hearing loss, and were unable to benefit from appropriate amplification. Each patient was implanted with the 22-electrode array inserted into the scala tympani, using the facial recess technique. Seven of the nine patients have functioning 22-channel systems, whereas one patient has a single-channel system and one had 14 electrodes inserted because of an unsuspected obstruction in the scala tympani. All patients are regular users of the device and none have been lost to follow-up. Seven patients have completed the prescribed Nucleus training program, and two patients are in the early stages of training. All nine patients have shown a restoration of hearing sensation in response to acoustic stimuli and a recognition of a wide variety of environmental sounds. All seven patients who have completed training and are using the multichannel stimulation have shown an improvement in their vowel and consonant recognition scores when the implant is used in conjunction with lipreading. Mean speech-tracking scores for these patients show an improvement from lipreading alone to lipreading with implant of 28.8 to 60.6 words per minute. Patients also demonstrated a consistent increased ability to use suprasegmental information and to obtain closed set work recognition on portions of the Minimal Auditory Capabilities test battery. Several of the patients have shown an ability to understand significant amounts of open set speech without lipreading. Two patients can comprehend noncoded telephone conversation; one scores an average of 42% on open set speech discrimination testing and the other 20% using the W22 word list with audition only.

Visual and tactual perception of syllable number in sentence stimuli.

The purpose of this study was to determine whether suprasegmental information could be perceived by artificially deafened subjects via the Multipoint Electrotactile Speech Aid (MESA) so as to enhance the visual perception of connected discourse. This study compared subjects' untrained ability to perceive the number of syllables in sentence stimuli in three conditions: Visual Alone (VA), Tactile Alone (TA), and Tactile plus Visual (TV). Five adults tapped the temporal pattern of stimuli using a telegraph key. Stimuli consisted of 1-, 3-, 5-, and 7-syllable meaningful sentences and 5- and 7-syllable (nonmeaningful) second-order sentences. The subjects perceived correctly a mean of 37% of the stimuli in the TA condition, and approximately 60% of the stimuli in the VA and TV conditions. Three possible explanations are offered to account for these results: (1) the limitations of the MESA's electrocutaneous transform of the speech signal, (2) the sensory limitations of the skin, and (3) the absence of extensive training.

Voice pitch as an aid to lipreading.

The totally deafened adult, unable to make use of a hearing aid, has no alternative to lipreading for everyday communication. Lipreading, however, is no substitute for hearing speech. Many lipreaders have great difficulty in ideal conditions and even the best lipreaders find the task demanding and tiring. Prosthetic attempts to substitute for lost hearing have centred on three distinct types of intervention, visual, tactile, and electrocochlear. As none of these is likely to yield a good understanding of a speech independent of lipreading in the near future, we have attempted to isolate relatively simple patterns of stimulation that, although not intelligible in themselves, well aid lipreading. From this point of view, the fundamental frequency or 'pitch' of the voice is the most important pattern element because if provides both segmental and suprasegmental information and is practically invisible. It thus complements the visual information already available on the face. As we show here, with the voice pitch presented acoustically, normal listeners can lipread a speaker reading continuous text at up to two and a half times the rate possible on the basis of lipreading alone. The pitch signal by itself, of course, is completely unintelligible. Although our work is primarily concerned with methods of electrical stimulation of the cochlea, it has implications for other sensory substitution techniques, the design of special purpose hearing aids and current theories of speech perception.

Organization in the perception of segmental and suprasegmental information by infants

In this study, we investigated whether infants are capable of discriminating not only differences in pitch and syllable-initial consonants but also the manner in which these components are combined. Using a modification of the high-amplitude sucking procedure, we found that three- and four-month-old infants were able to recognize rearrangements of these components, which, together with earlier findings, indicates that the infant's speech processing system is sensitive to the organization in the basic elements of human language.

Interactions in processing segmental and suprasegmental features of speech.

A speeded classification task was used to determine the nature of the dependency relation that exists during the processing of a segmental distinction in vowel quality and a suprasegmental distinction in pitch (Experiment 1) or loudness (Experiment 2) in a consonantvowel syllable. In both experiments, evidence was found for a mutual, symmetrical dependency between the processes underlying the analysis of the segmental and suprasegmental information in the syllable. This type of interaction pattern contrasts with the pattern of unidirectional dependency previously found for the analysis of a consonantal distinction and a suprasegmental distinction, in which the analysis of the consonantal information was found to be dependent on the analysis of the suprasegmental information (e.g., Wood, 1974). Together with these earlier findings, the present results clearly indicate that the form of the interaction between processes responsible for segmental and suprasegmental analysis is a function of the type of segmental information being analyzed. Future research must determine whether the distinct interaction patterns found thus far for consonants and vowels are due to a difference in phonetic class per se, i.e., consonant vs. vowel, or to a difference in the nature of the acoustic information specifying the consonant and vowel.