Organization in infant speech perception.

Over the past few years it has become increasingly clear that infants are capable of distinguishing speech contrasts that are differentiated by a single acoustic parameter and, moreover, that they perceive these distinctions in terms of adult phonetic categories. Three experiments with 200 2–4 mo olds examined whether infants also perceive organization in speech, specifically whether they are sensitive not only to individual elements, but also to the combination of elements that comprise basic units. Data show that Ss perceived organization at both the syllabic and segmental levels: Ss noticed the rearrangement of consonants and vowels to form new syllables and the rearrangement of phonetic feature values to form new phonetic segments. These findings provide further support for the assumption that very young infants have quite sophisticated speech perception abilities that constitute an important prerequisite for the acquisition of language. (French abstract) (37 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

An implantable electromagnetic sound source for speech production

Many operative and nonsurgical procedures have been designed for vocal rehabilitation following laryngectomy, but as yet a consistently successful solution with a negligible complication rate has failed to materialize. In our laboratory during the last two years, experiments have been carried out in dogs using active electromagnetic devices placed in the retropharynx as an energy source for speech production. An implantable porous coated stainless steel electromagnetic device energized from an externally controlled oscillator has been developed. Factors under investigation included biocompatibility of materials, durability, noise quality and intensity, together with the efficiency of the system. As a result of the favorable findings, it is envisaged that this system will soon be used in a carefully designed restricted clinical trial as a secondary procedure in those consenting cured laryngeal cancer patients who have failed to develop esophageal speech.     

Mechanisms of phonological inference in speech perception.

Cross-modal priming experiments have shown that surface variations in speech are perceptually tolerated as long as they occur in phonologically viable contexts. For example, [fre[i]p] (frayp) gains access to the mental representation of freight when in the context of [fre[i]pbeara] (frayp bearer) because the change occurs in normal speech as a process of place assimilation. The locus of these effects in the perceptual system was examined. Sentences containing surface changes were created that either agreed with or violated assimilation rules. The lexical status of the assimilated word also was manipulated, contrasting lexical and nonlexical accounts. Two phoneme monitoring experiments showed strong effects of phonological viability for words, with weaker effects for nonwords. It is argued that the listener's percept of the form of speech is a product of a phonological inference process that recovers the underlying form of speech. This process can operate on both words and nonwords, although it interacts with the retrieval of lexical information. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Hemiface contributions to hemispheric dominance in visual speech perception.

Lateralized displays are used widely to investigate hemispheric asymmetry in language perception. However, few studies have used lateralized displays to investigate hemispheric asymmetry in visual speech perception, and those that have yielded mixed results. This issue was investigated in the current study by presenting visual speech to either the left hemisphere (LH) or the right hemisphere (RH) using the face as recorded (normal), a mirror image of the normal face (reversed), and chimeric displays constructed by duplicating and reversing just one hemiface (left or right) to form symmetrical images (left-duplicated, right-duplicated). The projection of displays to each hemisphere was controlled precisely by an automated eye-tracking technique. Visual speech perception showed the same, clear LH advantage for normal and reversed displays, a greater LH advantage for right-duplicated displays, and no hemispheric difference for left-duplicated displays. Of particular note is that perception of LH displays was affected greatly by the presence of right-hemiface information, whereas perception of RH displays was unaffected by changes in hemiface content. Thus, when investigated under precise viewing conditions, the indications are not only that the dominant processes of visual speech perception are located in the LH but that these processes are uniquely sensitive to right-hemiface information. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Does lexical information influence the perceptual restoration of phonemes?

A critical issue in modeling speech perception is whether lexical representations can affect lower level (e.g., phonemic) processing. Phonemic restoration studies have provided support for such top-down effects, but there have also been a number of failures to find them. A methodology is introduced that provides good approximations to the underlying distributions of perceived intactness that are assumed in signal detection analyses of restoration. This methodology provides a sensitive means to determine the necessary conditions for lexical feedback to occur. When these conditions are created, a reliable lexical influence on phonemic perception results. The experiments thus show that lexical activation does influence lower level processing, and that these influences are fragile. The theoretical implications of real but fragile lexical effects are discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Discomforting talk and speech disruptions.

Laboratory studies over the past 20 yrs have shown a general relationship between a speaker's discomfort and the occurrence of speech disruptions, but critical task variables have not been clear. The present study examined the S's speech disruptions in describing 4 different experiences that had once made the S angry. 62 undergraduates served as Ss. By asking about different targets of anger (e.g., a family member), the interviewer varied the sensitivity of the episode. Various measures showed that the 4 tales varied in their sensitivity, and the results show increasingly more speech disruptions for more sensitive tales. Implications for psychotherapeutic mechanisms are discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The development of gesture and speech as an integrated system

Children, even at the one-word stage of language development, spontaneously produce gestures along with their speech, just as adults do. Although there appears to be a brief period prior to the onset of two-word speech during which gesture and speech do not form a well-integrated system, the ability to coordinate gesture and speech to convey a single message--and to "read" others' gestures with their speech to comprehend a message--develops early and is maintained throughout life. Gesture-speech combinations deliver a coherent message to the listener despite the fact that they consist of two different modalities of expression. According to McNeill (1992; Chapter One), this coherence is possible because gesture and speech share a common cognitive representation; that is, before the communication unfolds, gesture and speech are part of a single idea. As expression proceeds the message is parsed, with most information channeled into speech but some information channeled into gesture. Speech conveys information in a segmented, combinatorial format, whereas gesture conveys information in a global, mimetic format (see Goldin-Meadow, McNeill, and Singleton, 1996). Thus gesture and speech need not, and in fact often do not, convey the same information within a single utterance. Because gesture and speech form a unified system, mismatches between them can be a source of insight into the cognitive state of the speaker. And, indeed, it turns out that in both the young, one-word speaker and the older child (and possibly adults as well; Perry and Elder, 1996), a difference--or mismatch--between the information conveyed in gesture and the information conveyed in speech can signal readiness for cognitive growth. Whether the actual production of gesture-speech mismatches contributes to cognitive growth is an open question. That is, does the act of expressing two different pieces of information across modalities but within a single communicative act improve a child's ability to transpose that knowledge to a new level and thus express those pieces of information within a single modality? More work is needed to investigate whether the act of producing gesture-speech mismatches itself facilitates transition. Even if it turns out that the production of gesture-speech mismatches has little role to play in facilitating cognitive change, mismatch remains a reliable marker of the speaker's potential for cognitive growth. As such, an understanding of the relationship between gesture and speech may prove useful in clinical settings. For example, there is some evidence that children with delayed onset of two-word speech fall naturally into two groups: children who eventually achieve two-word speech, albeit later than the norm (that is, late bloomers), and children who continue to have serious difficulties with spoken language and may never be able to combine words into a single string (Feldman, Holland, Kemp, and Janosky, 1992; Thal, Tobias, and Morrison, 1991). Observation of combinations in which gesture and speech convey different information may prove a useful clinical tool for distinguishing, at a relatively young age, children who will be late bloomers from those who will have great difficulty mastering spoken language without intervention (see Stare, 1996, for preliminary evidence that the relationship between gesture and speech in children with unilateral brain damage correlates with early versus late onset of two-word combinations. In sum, for both speakers and listeners, gesture and speech are two aspects of a single process, with each modality contributing its own unique level of representation. Gesture conveys information in the global, imagistic form for which it is well suited, and speech conveys information in the segmented, combinatorial fashion that characterizes linguistic structures. The total representation of any message is therefore a synthesis of the analog gestural mode and the discrete speech mode. (ABSTRACT TRUNCATED)     

Prolonged speech and modification of stuttering: perceptual, acoustic, and electroglottographic data

Prolonged speech and its variants are a group of novel speech patterns that form the basis of a popular treatment for stuttering (Ingham, 1984). It is difficult to determine which features of prolonged speech are necessary for the elimination of stuttered speech because the speech pattern produces simultaneous changes in respiratory, laryngeal, and articulatory activity. Experimental studies have shown that the modification of phonation and of speech rate contributes to stuttering reduction, and increased duration of speech segments and reduced variability of vowel duration are known to occur as a result of prolonged-speech treatment programs. However, previous studies of prolonged speech have all instructed subjects to modify their customary speech patterns in a particular way. The aim of the present study was to investigate changes in the speech pattern of individual subjects when stuttering was modified with prolonged speech without specific instruction in how this should be done. In one experimental phase, 3 subjects showed clinically significant stuttering reductions when instructed to use whichever features of prolonged speech they needed to reduce their stuttering. The resulting perceptually stutter-free speech was judged to be natural sounding. Stuttering in a fourth subject reduced without experimental intervention. Recordings of acoustic and electroglottographic signals from the 4 subjects were analyzed. Changes in the variability of vowel duration occurred in all subjects. Theoretical and clinical implications of the results are discussed.     

A parallel licensing model of normal slips and phonemic paraphasias

A model of phonological processing in speech production based on prosodic licensing can capture general patterns of errors found in both normal and aphasic speech. All segments must be licensed by some prosodic category (syllable, nucleus, or rime) in order to be produced. Constraints on licensing, including both phonotactic and binding constraints, ensure that only correct licensing associations are retained. A computer simulation of our model produces utterances in qualitative agreement with human speech error data. Phonemic puraphasias are claimed to arise from the same mechanisms as normal speech errors; the difference being only a matter of disturbance of the lexical retrieval and licensing processes. The fact that these errors, which can involve gross disruption of the segmental sequence, still produce phonotactically well-formed strings is a direct consequence of the syllabic licensing that forms the core of our theory of speech production.     

Error biases in inner and overt speech: Evidence from tongue twisters.

To compare the properties of inner and overt speech, Oppenheim and Dell (2008) counted participants' self-reported speech errors when reciting tongue twisters either overtly or silently and found a bias toward substituting phonemes that resulted in words in both conditions, but a bias toward substituting similar phonemes only when speech was overt. Here, we report 3 experiments revisiting their conclusion that inner speech remains underspecified at the subphonemic level, which they simulated within an activation-feedback framework. In 2 experiments, participants recited tongue twisters that could result in the errorful substitutions of similar or dissimilar phonemes to form real words or nonwords. Both experiments included an auditory masking condition, to gauge the possible impact of loss of auditory feedback on the accuracy of self-reporting of speech errors. In Experiment 1, the stimuli were composed entirely from real words, whereas, in Experiment 2, half the tokens used were nonwords. Although masking did not have any effects, participants were more likely to report substitutions of similar phonemes in both experiments, in inner as well as overt speech. This pattern of results was confirmed in a 3rd experiment using the real-word materials from Oppenheim and Dell (in press). In addition to these findings, a lexical bias effect found in Experiments 1 and 3 disappeared in Experiment 2. Our findings support a view in which plans for inner speech are indeed specified at the feature level, even when there is no intention to articulate words overtly, and in which editing of the plan for errors is implicated. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Irrelevant speech and irrelevant tones: The relative importance of speech to the irrelevant speech effect.

Irrelevant auditory stimuli disrupt immediate serial recall. In the equipotentiality hypothesis, D. M. Jones and W. J. Macken (see record 1993-20312-001) made the controversial prediction that speech and tones have an equivalent disruptive effect. In the present study, 5 experiments tested their hypothesis. Experiments 1–4 showed that meaningful speech disrupts recall more than do tones. Experiments 3 and 4 provided some evidence that meaningful speech disrupts recall more than does meaningless speech, and Experiment 4 showed that even meaningless speech disrupts recall more than do tones. Using slightly different experimental procedures, Experiment 5 showed that letters disrupt recall more than do tones. Implications of these results for a number of theories of primary memory and the irrelevant speech effect are discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Ambiguity in speech

Ambiguity in speech during analysis is one manifestation of transference. Four types of ambiguity are illustrated. Each ambiguity communicates an unconscious wish or intention and the defense against this wish or intention. In each type of ambiguity, transference manifestations are expressed in the form of speech and not alone by its content, as we are accustomed to identifying the transference. Each type of ambiguity within the analysis expresses fear to responsibility and recrimination for libidinal or aggressive intentions. In some patients, the combinative ambiguity or malapropism expresses the fear of reprisal through the transformation into a self-inflicted injury. The pronominal ambiguity may express a feared and wished-for fusion with the analyst. The analyst's ambiguous interpretations are properly used to encourage associations, but at times may reflect lack of understanding and be a manifestation of countertransference.     

Cepstral representation of speech motivated by time-frequency masking: an application to speech recognition

A new spectral representation incorporating time-frequency forward masking is proposed. This masked spectral representation is efficiently represented by a quefrency domain parameter called dynamic-cepstrum (DyC). Automatic speech recognition experiments have demonstrated that DyC powerfully improves performance in phoneme classification and phrase recognition. This new spectral representation simulates a perceived spectrum. It enhances formant transition, which provides relevant cues for phoneme perception, while suppressing temporally stationary spectral properties, such as the effect of microphone frequency characteristics or the speaker-dependent time-invariant spectral feature. These features are advantageous for speaker-independent speech recognition. DyC can efficiently represent both the instantaneous and transitional aspects of a running spectrum with a vector of the same size as a conventional cepstrum. DyC is calculated from a cepstrum time sequence using a matrix lifter. Each column vector of the matrix lifter performs spectral smoothing. Smoothing characteristics are a function of the time interval between a masker and a signal. DyC outperformed a conventional cepstrum parameter obtained through linear predictive coding (LPC) analysis for both phoneme classification and phrase recognition by using hidden Markov models (HMMs). Compared with speaker-dependent recognition, an even greater improvement over the cepstrum parameter was found in speaker-independent speech recognition. Furthermore, DyC with only 16 coefficients exhibited higher speech recognition performance than a combination of the cepstrum and a delta-cepstrum with 32 coefficients for the classification experiment of phonemes contaminated by noises.     

Articulatory duration in single-word speech production.

Three different speech production paradigms assessed C. T. Kello, D. C. Plaut, and B. MacWhinney's (2000) claim that the characteristics of speech production flexibly vary between staged and cascaded modes depending on task demand. All experiments measured response latencies and durations of single words without and with a response deadline. Experiment 1 used a picture-word interference task; Experiment 2 blocked pictures either by semantic category or by word-initial overlap; and Experiment 3 used a Stroop paradigm. In all cases, systematic effects of semantic and form relatedness were obtained on latencies but not on response durations. These results support the assumption that articulation, as assessed by response duration, is never influenced by central cognitive processes once a response has been initiated. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Auditory and auditory-visual perception of clear and conversational speech

Research has shown that speaking in a deliberately clear manner can improve the accuracy of auditory speech recognition. Allowing listeners access to visual speech cues also enhances speech understanding. Whether the nature of information provided by speaking clearly and by using visual speech cues is redundant has not been determined. This study examined how speaking mode (clear vs. conversational) and presentation mode (auditory vs. auditory-visual) influenced the perception of words within nonsense sentences. In Experiment 1, 30 young listeners with normal hearing responded to videotaped stimuli presented audiovisually in the presence of background noise at one of three signal-to-noise ratios. In Experiment 2, 9 participants returned for an additional assessment using auditory-only presentation. Results of these experiments showed significant effects of speaking mode (clear speech was easier to understand than was conversational speech) and presentation mode (auditory-visual presentation led to better performance than did auditory-only presentation). The benefit of clear speech was greater for words occurring in the middle of sentences than for words at either the beginning or end of sentences for both auditory-only and auditory-visual presentation, whereas the greatest benefit from supplying visual cues was for words at the end of sentences spoken both clearly and conversationally. The total benefit from speaking clearly and supplying visual cues was equal to the sum of each of these effects. Overall, the results suggest that speaking clearly and providing visual speech information provide complementary (rather than redundant) information.     

Enhancement of electrolaryngeal speech by adaptive filtering

Artificial larynges provide a means of verbal communication for people who have either lost or are otherwise unable to use their larynges. Although they enable adequate communication, the resulting speech has an unnatural quality and is significantly less intelligible than normal speech. One of the major problems with the widely used Transcutaneous Artificial Larynx (TAL) is the presence of a steady background noise caused by the leakage of acoustic energy from the TAL, its interface with the neck, and the surrounding neck tissue. The severity of the problem varies from speaker to speaker, partly depending upon the characteristics of the individual's neck tissue. The present study tests the hypothesis that TAL speech is enhanced in quality (as assessed through listener preference judgments) and intelligibility by removal of the inherent, directly radiated background signal. In particular, the focus is on the improvement of speech over the telephone or through some other electronic communication medium. A novel adaptive filtering architecture was designed and implemented to remove the background noise. Perceptual tests were conducted to assess speech, from two individuals with a laryngectomy and two normal speakers using the Servox TAL, before and after processing by the adaptive filter. A spectral analysis of the adaptively filtered TAL speech revealed a significant reduction in the amount of background source radiation yet preserved the acoustic characteristics of the vocal output. Results from the perceptual tests indicate a clear preference for the processed speech. In general, there was no significant improvement or degradation in intelligibility. However, the processing did improve the intelligibility of word-initial non-nasal consonants.     

Cognitive control factors in speech perception at 11 months.

The development of speech perception during the 1st year reflects increasing attunement to native language features, but the mechanisms underlying this development are not completely understood. One previous study linked reductions in nonnative speech discrimination to performance on nonlinguistic tasks, whereas other studies have shown associations between speech perception and vocabulary growth. The present study examined relationships among these abilities in 11-month-old infants using a conditioned head-turn test of native and nonnative speech sound discrimination, nonlinguistic object-retrieval tasks requiring attention and inhibitory control, and the MacArthur-Bates Communicative Development Inventory (L. Fenson et al., 1993). Native speech discrimination was positively linked to receptive vocabulary size but not to the cognitive control tasks, whereas nonnative speech discrimination was negatively linked to cognitive control scores but not to vocabulary size. Speech discrimination, vocabulary size, and cognitive control scores were not associated with more general cognitive measures. These results suggest specific relationships between domain-general inhibitory control processes and the ability to ignore variation in speech that is irrelevant to the native language and between the development of native language speech perception and vocabulary. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Case-control study of prenatal ultrasonography exposure in children with delayed speech

OBJECTIVE: To determine whether there is an association between prenatal ultrasound exposure and delayed speech in children. DESIGN: Case-control study. SETTING: Network of community physicians affiliated with the Primary Care Research Unit, University of Calgary. SUBJECTS: Thirty-four practitioners identified 72 children aged 24 to 100 months who had undergone a formal speech-language evaluation and were found to have delayed speech of unknown cause by a speech-language pathologist. For each case subject the practitioners found two control subjects matched for sex, date of birth, sibling birth order and associated health problems. MAIN OUTCOME MEASURES: Rates of prenatal ultrasound exposure and delayed speech. RESULTS: The children with delayed speech had a higher rate of ultrasound exposure than the control subjects. The findings suggest that a child with delayed speech is about twice as likely as a child without delayed speech to have been exposed to prenatal ultrasound waves (odds ratio 2.8, 95% confidence limit 1.5 to 5.3; p = 0.001). CONCLUSION: An association between prenatal ultrasonography exposure and delayed speech was found. If there is no obvious clinical indication for diagnostic in-utero ultrasonography, physicians might be wise to caution their patients about the vulnerability of the fetus to noxious agents.     

Intelligibility of tracheoesophageal speech among naive listeners

Tracheoesophageal (TE) speech is now the most common method of voice rehabilitation after total laryngectomy. The speech intelligibility of laryngectomees who use TE speech as their primary mode of communication was evaluated by 20 "naive" listeners. Two speech intelligibility tests were administered using phonetically balanced rhyming words or lists of spondee words. The overall intelligibility for the group of laryngectomees was 76%, with a wide range of variability among the individual TE speakers. We concluded that TE speech is significantly less intelligible to naive listeners than normal laryngeal speech; further refinement of voice rehabilitation for laryngectomees is needed.     

Neural coding of relational invariance in speech: Human language analogs to the barn owl.

The ability to form perceptual equivalence classes from variable input stimuli is common in both animals and humans. Neural circuitry that can disambiguate ambiguous stimuli to arrive at perceptual constancy has been documented in the barn owl's inferior colliculus where sound-source azimuth is signaled by interaural phase differences spanning the frequency spectrum of the sound wave. Extrapolating from the sound-localization system of the barn owl to human speech, 2 hypothetical models are offered to conceptualize the neural realization of relative invariance in (a) categorization of stop consonants /b, d, g/ across varying vowel contexts and (b) vowel identity across speakers. 2 computational algorithms employing real speech data were used to establish acoustic commonalities to form neural mappings representing phonemic equivalence classes in the form of functional arrays similar to those seen in the barn owl. (PsycINFO Database Record (c) 2010 APA, all rights reserved)