Trained Listeners Research Articles

Hypotheses, design, and instruments in the Educatel Study, Brazil, 2015/2016.

The Educatel Study 2015/2016 was designed to evaluate health and work conditions in a representative sample of the 2,220,000 schoolteachers working in Basic Education in Brazil. The article aimed to describe the telephone survey's basis and design, using a questionnaire consisting of 54 short, simple questions, most of which with multiple-choice answers (closed questions) addressing diseases, accidents, absenteeism, frequency of healthy behaviors, physical and psychosocial environment, and employment characteristics. In the pilot stage, the multi-theme questionnaire was assessed in order to verify the effects of the terminology, the format of the questions and the multiple-choice answers, the questionnaire's internal organization, production of the answers, and duration of the interview. The interviewers' training and follow-up and listening of the calls in real time aimed to identify communications problems. The teachers were interviewed at school after prior contact with the school assistant to set appointments. The advantages and risks of biases related to the telephone interview modality should be considered to interpret the results. The results on the teachers' profile, illnesses, and school environment will provide inputs for elaborating inter-sector measures to improve the target group's health, which is related to Brazil's school system indicators based on the concepts presented here.

Quality Analysis of Streaming Audio over Mobile Networks

This paper utilizes open source software to analyze the quality of audio streamed over mobile cellular networks. Industry conventions have been developed to assess audio quality such as the Mean Opinion Score or MOS scale described in the International Telecommunications Union ITU P.800.1 document. The MOS scale is a subjective assessment based on the listener’s experience. To eliminate the use of a trained audio listener we automate an estimated MOS calculation by measuring the packet loss, average latency and jitter over the network transport path. The network under test is a pilot network to replace the dedicated analog circuit from the broadcast center to a radio transmitter. We intend to automate the logging of an objective quality assessment using MOS, cellular router and decoder measurements with confirmation using automatically generated visual representations of sample audio received. These visual representations will aid in manual confirmation of poor MOS scores with audio samples available for more in-depth review.

Perceptual-learning evidence for inter-onset-interval- and frequency-specific processing of fast rhythms.

Rhythm is fundamental to music and speech, yet little is known about how even simple rhythmic patterns are processed. Here we investigated the processing of isochronous rhythms in the short inter-onset-interval (IOI) range (IOIs < 250-400 ms) using a perceptual-learning paradigm. Trained listeners (n=8) practiced anisochrony detection with a 100-ms IOI marked by 1-kHz tones, 720 trials per day for 7 days. Between pre- and post-training tests, trained listeners improved significantly more than controls (no training; n=8) on the anisochrony-detection condition that the trained listeners practiced. However, the learning on anisochrony detection did not generalize to temporal-interval discrimination with the trained IOI (100 ms) and marker frequency (1 kHz) or to anisochrony detection with an untrained marker frequency (4 kHz or variable frequency vs. 1 kHz), and generalized negatively to anisochrony detection with an untrained IOI (200 ms vs. 100 ms). Further, pre-training thresholds were correlated among nearly all of the conditions with the same IOI (100-ms IOIs), but not between conditions with different IOIs (100-ms vs. 200-ms IOIs). Thus, it appears that some task-, IOI-, and frequency-specific processes are involved in fast-rhythm processing. These outcomes are most consistent with a holistic rhythm-processing model in which a holistic "image" of the stimulus is compared to a stimulus-specific template.

Can Listeners Hear Who Is Singing? The Role of Listener Experience in Singer Discrimination Across Pitch

Previous research has shown that listeners are unable to identify who is singing across pitch when the voices are unfamiliar to them. Implementing a very short training period, however, greatly improves this ability, but only when the voices being compared are of different voice category. The objective of this study was to determine whether experienced listeners with highly developed knowledge of voice categories can more easily discriminate between singers across pitch. This study used an ABX paradigm where listeners heard two different singers singing "ah" at the same pitch. Listeners identified which of the two singers produced a third "ah" at a different pitch. Stimuli were recorded from two baritones, two tenors, two mezzo-sopranos, and two sopranos across a 1.5 octave range. Data were collected from 42 inexperienced listeners and 27 experienced listeners. Experienced listeners were better at singer discrimination across pitch than were inexperienced listeners for all conditions except same-category comparisons at the interval of the third. Experienced listeners were better at singer discrimination across pitch than were slightly trained listeners for all conditions except same-category female singers at all pitch intervals. The ability to discriminate singers across pitch is the greatest for experienced listeners, followed by slightly trained inexperienced listeners, followed by inexperienced listeners.

A Phonetic Complexity-Based Approach for Intelligibility and Articulatory Precision Testing: A Preliminary Study on Talkers With Amyotrophic Lateral Sclerosis.

This study describes a phonetic complexity-based approach for speech intelligibility and articulatory precision testing using preliminary data from talkers with amyotrophic lateral sclerosis. Eight talkers with amyotrophic lateral sclerosis and 8 healthy controls produced a list of 16 low and high complexity words. Sixty-four listeners judged the samples for intelligibility, and 2 trained listeners completed phoneme-level analysis to determine articulatory precision. To estimate percent intelligibility, listeners orthographically transcribed each word, and the transcriptions were scored as being either accurate or inaccurate. Percent articulatory precision was calculated based on the experienced listeners' judgments of phoneme distortions, deletions, additions, and/or substitutions for each word. Articulation errors were weighted based on the perceived impact on intelligibility to determine word-level precision. Between-groups differences in word intelligibility and articulatory precision were significant at lower levels of phonetic complexity as dysarthria severity increased. Specifically, more severely impaired talkers showed significant reductions in word intelligibility and precision at both complexity levels, whereas those with milder speech impairments displayed intelligibility reductions only for more complex words. Articulatory precision was less sensitive to mild dysarthria compared to speech intelligibility for the proposed complexity-based approach. Considering phonetic complexity for dysarthria tests could result in more sensitive assessments for detecting and monitoring dysarthria progression.

Social and Stylistic Correlates of Vocal Fry in a cappella Performances

To determine the social and stylistic correlates of vocal fry in a cappella performances. A matched-guise experiment was used to measure listener evaluations of fry and non-fry guises. Four singers, two male and two female, sang "The Star-Spangled Banner" with onset vocal fry. These recordings were used to create the two guises: (i) an unmodified recording with onset vocal fry on vowel-initial words and (ii) a recording in which the fry had been removed. In total, 253 participants listened to the recordings and evaluated the singers' social and stylistic attributes along a Likert scale, e.g., how confident, sexy, and sincere each singer sounded. A factor analysis was used to conflate correlated variables, and mixed effects linear regression models (n = 1,012) were fitted to each lone or joint factor to determine whether vocal fry significantly influenced listeners' responses to the singers. Vocal fry significantly altered listener evaluations of the singers' sincerity/commitment, maturity/sophistication, naturalness, and confidence (P < 0.05). Unlike male singers, who were rated as significantly less sincere/committed with vocal fry, female singers were seen as more sincere/committed with vocal fry and younger listeners also found them less natural, suggesting vocal fry is associated with emotional intensity in female voices. Younger listeners perceived singers with fry as less mature/sophisticated, suggesting an association with youth. Finally, listeners with more musical training rated singers with fry as less confident, while less trained listeners did not exhibit this difference. Listeners are highly attuned to vocal fry in music but respond to it differently based upon their age, musical training, and the singer's sex. Vocal fry is evaluated more positively among younger, less musically trained listeners, and it is better received in women's voices, suggesting that the use of fry strategically targets a specific audience, i.e., younger and less trained listeners, who interpret fry as a marker of youth and emotional earnestness. These findings show that a single stylistic feature like vocal fry can be imbued with multiple meanings depending on the singer and audience, and its use can serve to include or exclude particular listener groups.

The boundaries of auditory perception for within-word syllable segregation in untrained and trained adult listeners

ABSTRACTSyllable segregation is among the core diagnostic features of both childhood apraxia of speech and acquired apraxia of speech; however, little is known about the limen of perception of syllable segregation. The purpose of this research was therefore to explore adult listeners’ auditory perception of within-word syllable segregation in trained and untrained adult listeners.Two experimental design studies, each with two phases, were conducted. Study one included 40 untrained listeners (aged 18–28 years), and study two included 5 trained listeners (10–25 years of experience). Recorded audio samples of multisyllabic non-words were manipulated to insert gaps of increasing length to simulate syllable segregation. All stimuli were anchored against an unaltered control sample. In each phase, 32 stimuli were randomly presented twice in a free-field listening task with listeners making Yes/No decisions to record their perception of segregation.At a 90% accuracy threshold, the untrained listeners’ limen was 90ms, while the trained listeners’ was 85ms. At an 80% accuracy threshold, both sets of listeners had a limen of 80ms. Overall, there were no significant differences in accuracy between the two listening groups. Gap duration was positively correlated with accurate perception across both studies. Both groups demonstrated good intra-rater reliability and excellent inter-rater reliability, with no significant differences between the untrained and trained listeners.These findings have implications for developing a standardised criteria for rating syllable segregation in clinical contexts. No evidence was found for the hypothesis that training makes a difference in listeners’ accuracy or reliability when rating syllable segregation.

Can Listeners Hear Who Is Singing? The Development of Voice Category Perception

This study sought to investigate if a small amount of training in identification of voices elicits the development of prototypical vocal categories. This study used a between-group design. This study used an ABX paradigm where listeners heard two different singers singing "ah" at the same pitch. Listeners identified which of the two singers was the producer of a third "ah" at a different pitch. Stimuli were recorded from two baritones, two tenors, two mezzo-sopranos, and two sopranos across a 1.5-octave range. Participants were randomly assigned to two groups, one group that received a training session using voices that were different from those in the experimental session, but of the same voice category, and one group that received no training. Training listeners with voices that are different from those of singers presented in the experiment did not significantly improve the ability to discriminate individual voices of the same voice category, but did significantly improve that ability to discriminate individual voices when the voices being compared were of different voice categories. Small amounts of purposeful exposure to human voices appear to result in the beginnings of listener voice category formation, providing listeners with prototypical categories that can aid them in discrimination of novel voices of those same categories.

Using high variability phonetic training to train non-tonal listeners with no musical background to perceive lexical tones

Previous research has not extensively investigated whether High Variability Phonetic Training (HVPT) is effective in training listeners with no musical background and no prior experience with a tone language in their identification of non-native lexical tones. In this study, it was investigated whether HVPT is applicable to the acquisition of non-native tones by such listeners. Twenty-one speakers of American English were trained in eight sessions using the HVPT approach to identify Mandarin tones in monosyllabic words. Ten of the participants were exposed to words produced by multiple talkers (MT condition), and eleven participants were exposed to words produced by a single talker (ST condition). The listeners’ identification accuracy revealed an average 44% increase from the pretest to the posttest for the MT condition and an average 30% increase for the ST condition. The improvement also generalized to new monosyllabic words produced by a familiar talker and those produced by two unfamiliar talkers. The learning however did not generalize to novel disyllabic words produced either by a familiar talker or an unfamiliar talker. Comparisons between two groups further revealed that the improvement of the listeners in the MT condition was significantly higher than that of the listeners in the ST condition.

Cross-sectional follow-up of voice outcomes in children who have a history of airway reconstruction surgery.

This study reports vocal function in a cross-section of children with subglottic stenosis. Each child had a history of laryngotracheal reconstruction and/or cricotracheal resection surgery. Vocal function was measured using laryngoscopy, acoustic analysis, perceptual evaluation and impact of voice on quality of life. All patients aged >5 years with history of laryngotracheal reconstruction and/or cricotracheal resection surgery at the Scottish National Complex Airways service were invited to participate. Data were gathered in the Royal Hospital for Children in Glasgow in a single outpatient appointment. Twelve of 56 former patients (aged 5-27) provided a voice sample and eleven consented to awake laryngoscopy. All consented for detailed evaluation of their medical records. Acoustic analysis of fundamental frequency and pitch perturbation was conducted on sustained vowel [a]. Perceptual evaluation was conducted by 4 trained listeners on a series of spoken sentences. Impact on quality of life was measured using the paediatric voice-related quality of life questionnaire. Laryngeal function was descriptively evaluated. Four children had normal voice acoustically, perceptually and in relation to voice-related quality of life. One of these had vocal fold nodules unrelated to surgical history. Two other children had "near normal" vocal function, defined where most voice measurements fell within the normal range. Normal or "near normal" voice is a possible outcome for children who have had this surgery. Where there is an ongoing complex medical condition, voice outcome may be poorer.

Effect of control samples and listener attributes on speech naturalness ratings of people who stutter

PurposeSpeech restructuring treatment can effectively reduce stuttering but the resultant speech may sound unnatural. Martin et al. (1984) speech naturalness scale is widely used by clinicians and researchers, yet little is known about whether including normally fluent speech samples alters the judgement of the naturalness of speech samples of people who stutter, and whether attributes of listeners – specifically training and sex – influence ratings. MethodsIn this study 20 untrained listeners (male and female) and 19 speech language pathology students (female only) rated either the naturalness of 21 speech samples from adults who stutter obtained post-treatment, or the same 21 post-treatment samples randomly mixed with samples of 21 samples from normally fluent speakers matched for age and sex. The independent variables were sample composition (addition of fluent controls) and listener training. The dependent variable was listener naturalness rating. ResultsA two-factor ANOVA with listener training and sample composition as independent variables and naturalness ratings as the dependent variable was performed. Untrained listeners rated samples as significantly less natural than trained listeners. The addition of control samples did not significantly impact scores assigned to post-treatment samples. A comparison of male and female listeners was completed using the Mann Whitney U Test. A significant group difference was observed with female listeners rating the samples more leniently (more natural) than male listeners. ConclusionBased on this preliminary research, the addition of controls does not appear necessary in evaluating speech naturalness, however the composition of the listener group may affect results.

Pitch Dispersal and the Perception of Tonal Strength in Schoenberg’s Oeuvre

The idea that listeners’ tonal/atonal sense represents a special case of multiple causation was examined, and the following hypothesis was tested: pitch dispersal (i.e., distance in pitch between successive tones) is a secondary determiner of tonality and atonality, the former being strengthened by low levels of pitch dispersal and the latter by high levels of pitch dispersal. A correlational study was conducted in which eight trained listeners judged the degree to which 78 melodies extracted from A. Schoenberg’s oeuvre convey a tonic. In line with the present hypothesis, results suggest that listeners’ judgments were influenced not only by consonance or pitch class distributions (i.e., by underlying “scales”), as expected from previous research, but also by pitch dispersal. Interestingly, it was also found that Schoenberg’ melodies became not only less diatonic over time, but also more dispersed, which suggests that the joint manipulation of pitch class distributions and pitch dispersal might have been a strategy on Schoenberg’s part to weaken the sense of tonality. Some of the key musicological, theoretical, and psychological implications of these findings are discussed.

Selecting an acoustic correlate for automated measurement of /r/ production in children

A current need in the field of speech pathology is the development of reliable and efficient techniques for the purpose of evaluating changes in speech production over the course of treatment. The industry standard for scoring speech is time consuming and expensive, as it involves aggregating perceptual ratings across expert listeners. As techniques for automated measurement of speech improve, acoustic measures have the potential to play an expanded role in the clinical management of speech disorders. The current study asks which of several acoustic measures of children’s productions of English /r/ corresponds most closely with ratings given by trained listeners. This study fits a series of ordinal mixed effects regression models to a large sample of children’s /r/ productions that had previously been rated by three trained listeners (speech-language pathologists). Controlling for age, sex, and allophonic contextual differences, the acoustic measure that accounted for the most variance in speech rating was F3-F2 distance, normalized relative to a sample of age- and gender-matched speakers. Therefore, this acoustic measure is recommended for use in future automated scoring of children’s production of rhotic targets.

Context-based Composition in an Interdisciplinary Collaborative Framework

This article describes two site-specific sound installations conceived as part of interdisciplinary collaborations which aimed to explore different acoustic and perceptual features of urban and rural landscapes. The creative process involved in the design and implementation of both installations is described, focusing on the organic montage techniques based on spatial and temporal features of field recording methods. Future developments of the works will consider ways of developing further spatial-temporal sampling methods to assess particular features of soundscapes by panels of trained listeners.

Attentional modulation and domain-specificity underlying the neural organization of auditory categorical perception.

Categorical perception (CP) is highly evident in audition when listeners' perception of speech sounds abruptly shifts identity despite equidistant changes in stimulus acoustics. While CP is an inherent property of speech perception, how (if) it is expressed in other auditory modalities (e.g., music) is less clear. Moreover, prior neuroimaging studies have been equivocal on whether attentional engagement is necessary for the brain to categorically organize sound. To address these questions, we recorded neuroelectric brain responses [event-related potentials (ERPs)] from listeners as they rapidly categorized sounds along a speech and music continuum (active task) or during passive listening. Behaviorally, listeners' achieved sharper psychometric functions and faster identification for speech than musical stimuli, which was perceived in a continuous mode. Behavioral results coincided with stronger ERP differentiation between prototypical and ambiguous tokens (i.e., categorical processing) for speech but not for music. Neural correlates of CP were only observed when listeners actively attended to the auditory signal. These findings were corroborated by brain-behavior associations; changes in neural activity predicted more successful CP (psychometric slopes) for active but not passively evoked ERPs. Our results demonstrate auditory categorization is influenced by attention (active>passive) and is stronger for more familiar/overlearned stimulus domains (speech>music). In contrast to previous studies examining highly trained listeners (i.e., musicians), we infer that (i) CP skills are largely domain-specific and do not generalize to stimuli for which a listener has no immediate experience and (ii) categorical neural processing requires active engagement with the auditory stimulus.

Training F0 discrimination improves vowel identification among speech m

Listeners can use voice or fundamental frequency (F0) differences to separate target speech from competing talkers. Discrimination of F0 of harmonic tones has been shown to improve with adaptive training. Here, we examine whether such F0 learning can benefit speech identification in noise. Normal-hearing adults practiced F0 discrimination with standard F0 roved between 120 and 240 Hz. Before and after training, these listeners and a no-contact control group were tested on a vowel identification task, in which Mandarin vowels were embedded within six-talker babble noise at multiple signal-to-noise ratios. After training, the trained listeners improved more than controls on F0 discrimination. Critically, they also improved more in vowel identification. Threshold SNR at 60% intelligibility dropped 1.9 dB as a result of F0 training. The results demonstrate that speech perception in noise can benefit from learning of basic auditory skills, revealing a promising approach for rehabilitative training.

The effects of exposure and training on the perception of time-compressed speech in native versus nonnative listeners.

The present study investigated the effects of language experience on the perceptual learning induced by either brief exposure to or more intensive training with time-compressed speech. Native (n = 30) and nonnative (n = 30) listeners were each divided to three groups with different experiences with time-compressed speech: A trained group who trained on the semantic verification of time-compressed sentences for three sessions, an exposure group briefly exposed to 20 time-compressed sentences, and a group of naive listeners. Recognition was assessed with three sets of time-compressed sentences intended to evaluate exposure-induced and training-induced learning as well as across-token and across-talker generalization. Learning profiles differed between native and nonnative listeners. Exposure had a weaker effect in nonnative than in native listeners. Furthermore, native and nonnative trained listeners significantly outperformed their untrained counterparts when tested with sentences taken from the training set. However, only trained native listeners outperformed naive native listeners when tested with new sentences. These findings suggest that the perceptual learning of speech is sensitive to linguistic experience. That rapid learning is weaker in nonnative listeners is consistent with their difficulties in real-life conditions. Furthermore, nonnative listeners may require longer periods of practice to achieve native-like learning outcomes.

Auditory Perceptual Learning in Adults with and without Age-Related Hearing Loss.

Introduction : Speech recognition in adverse listening conditions becomes more difficult as we age, particularly for individuals with age-related hearing loss (ARHL). Whether these difficulties can be eased with training remains debated, because it is not clear whether the outcomes are sufficiently general to be of use outside of the training context. The aim of the current study was to compare training-induced learning and generalization between normal-hearing older adults and those with ARHL.Methods : Fifty-six listeners (60–72 y/o), 35 participants with ARHL, and 21 normal hearing adults participated in the study. The study design was a cross over design with three groups (immediate-training, delayed-training, and no-training group). Trained participants received 13 sessions of home-based auditory training over the course of 4 weeks. Three adverse listening conditions were targeted: (1) Speech-in-noise, (2) time compressed speech, and (3) competing speakers, and the outcomes of training were compared between normal and ARHL groups. Pre- and post-test sessions were completed by all participants. Outcome measures included tests on all of the trained conditions as well as on a series of untrained conditions designed to assess the transfer of learning to other speech and non-speech conditions.Results : Significant improvements on all trained conditions were observed in both ARHL and normal-hearing groups over the course of training. Normal hearing participants learned more than participants with ARHL in the speech-in-noise condition, but showed similar patterns of learning in the other conditions. Greater pre- to post-test changes were observed in trained than in untrained listeners on all trained conditions. In addition, the ability of trained listeners from the ARHL group to discriminate minimally different pseudowords in noise also improved with training.Conclusions : ARHL did not preclude auditory perceptual learning but there was little generalization to untrained conditions. We suggest that most training-related changes occurred at higher level task-specific cognitive processes in both groups. However, these were enhanced by high quality perceptual representations in the normal-hearing group. In contrast, some training-related changes have also occurred at the level of phonemic representations in the ARHL group, consistent with an interaction between bottom-up and top-down processes.

Effect of stimulus bandwidth and duration on monaural envelope correlation perception.

Monaural envelope correlation perception is the ability to discriminate between stimuli composed of two or more bands of noise based on envelope correlation. Sensitivity decreases as stimulus bandwidth is reduced below 100 Hz. The present study manipulated stimulus bandwidth (25-100 Hz) and duration (25-800 ms) to evaluate whether performance of highly trained listeners is limited by the number of inherent modulation periods in each presentation. Stimuli were two bands of noise, separated by a 500-Hz gap centered on 2250 Hz. Performance improved reliably with increasing numbers of envelope modulation periods, although there were substantial individual differences.

Speech-related fatigue and fatigability in Parkinson's disease

This study tested the assumption that speech is more susceptible to fatigue than normal in persons with dysarthria. After 1 h of speech-like exercises, participants with Parkinson's disease (PD) were expected to report increased perceptions of fatigue and demonstrate fatigability by producing less precise speech with corresponding acoustic changes compared to neurologically normal participants. Twelve adults with idiopathic PD and 13 neurologically normal adults produced sentences with multiple lingual targets before and after six 10-min blocks of fast syllable or word productions. Both groups reported increasing self-perceived fatigue over time, but trained listeners failed to detect systematic differences in articulatory precision or speech naturalness between sentences produced before and after speech-related exercises. Similarly, few systematic acoustic differences occurred. These findings do not support the hypothesis that dysarthric speakers are particularly susceptible to speech-related fatigue; instead, speech articulation generally appears to be resistant to fatigue induced by an hour of moderate functional exercises.