Speech Intelligibility In Speakers Research Articles

Impact of perceptual-learning strategies and background noise on disordered speech intelligibility

Objective: There is a plethora of research showing reduced speech intelligibility for a variety of voice disorders (i.e., dysphonia, alaryngeal). Therapeutic approaches to improve intelligibility typically involve targeting the speaker (e.g., clear speech, reduced rate) with minimal attention to the listener. Therefore, there were three purposes of this study: 1) to determine the impact of background noise on the speech intelligibility of disordered speakers; 2) to determine the impact of providing listeners with perceptual-learning strategies on the speech intelligibility of speakers with a voice disorder; and 3) to determine if subjective ratings of voice quality can predict speech intelligibility. Methods: Sentences were recorded from 12 speakers (2 typical, 3 alaryngeal, 7 dysphonic). Sentences were divided into one of three groups of signal-to-noise ratios (SNR: quiet, +5 dB SNR, and 0 dB SNR) and individually presented to 129 healthy listeners divided into one of three groups (i.e., control, acknowledgment of disorder, cognitive-perceptual strategies). Orthographic transcription was used to assess speech intelligibility. In addition, three expert listeners provided subjective voice quality ratings of all speakers. Results: Listeners had significantly more intelligibility errors with increased background noise (p <.001) and providing strategies to listeners did not result in a statistical improvement level F(6, 486) = 1.53, p = .17, η2p = 0.02. Regression analysis showed that the subjective voice quality overall severity was able to predict speech intelligibility in the noisy condition (0 dB SNR), accounting for 37% of the variance, R2 = .365, F(1,10) = 5.759, p = .037. Conclusion: Results suggest that increased background noise has a deleterious effect on the speech intelligibility of those with a voice disorder but that providing listeners with strategies in hopes of improving speaker intelligibility was not successful. Results did provide support, however, for the use of subjective voice quality ratings as a potential index of speech intelligibility.

Effects of multitasker babble noise on speech intelligibility in children with velopharyngeal insufficiency (VPI).

Intelligibility measurement is influenced by the characteristics of a speaker, listener and contextual factors. This study addresses the clinical problem of measuring speech intelligibility in children with velopharyngeal insufficiency (VPI) in real-world conditions. The purpose of the study was to investigate the effects of background noise on speech intelligibility in speakers with velopharyngeal insufficiency (VPI) compared to typical speech. The study further determined the contribution of nasalance and articulation accuracy in judgments of intelligibility. Fifteen speakers diagnosed with VPI and their typical peers provided audio recordings of 20 sentences from the Hearing in Noise Test. Speech samples were presented over quiet and noise (+5dB signal-to-noise ratio) conditions to 70 naïve listeners. Intelligibility scores from naïve listeners' orthographic transcriptions were obtained as the percentage of correctly identified words. A repeated-measures analysis of variance showed diagnosis of VPI (F(1, 28) = 13.44, p = 0.001, and presence of noise (F(1, 28) = 39.18, p < 0.001) significantly affected the intelligibility scores. There was no interaction between the diagnosis of VPI and noise (F(1, 28) = 0.06, p = 0.80). The multivariate regression analysis indicated that nasalance and articulation accuracy explain a significant amount of variance in the intelligibility scores of VPI speakers in quiet (F(2, 12) = 7.11, p < 0.05, R2 = 0.55, R2 Adjusted = 0.47) and noise (F(2, 12) = 6.32, p < 0.05, R2 = 0.51, R2 Adjusted = 0.43), but the significance mainly came from the effect of percentage of consonants correct (β = 0.97, t(12) = 2.90, p = 0.01). Percentage of consonants correct significantly increased the speech intelligibility in either with or without noise conditions. The current work suggests that background noise will significantly affect reductions in intelligibility in both groups; the effect is more prominent in VPI speech. It was also further noted that articulation accuracy significantly affected intelligibility in quiet and noise rather than nasalance scores. What is already known on the subject Intelligibility measurement is influenced by the characteristics of a speaker, listener and contextual factors. Accordingly, it is essential to determine the degree to which speech assessments in the clinic can predict communication difficulties in the presence of background noise in real life. Background noise can adversely cause speech intelligibility degradation in individuals with speech disorders. What this study adds The study examined the effects of background noise on speech intelligibility in speakers with velopharyngeal insufficiency (VPI) secondary to cleft palate compared to typical speech. The study results suggested that the presence of background noise will significantly affect reductions in intelligibility in both groups; however, the effect is more prominent in VPI speech. What are the clinical implications of this work? We found out that the intelligibility of VPI speech is lower in the presence of background noise, and therefore, assessments of speech intelligibility in clinical settings should take this into account. To ensure effective communication in noisy environments, recommended strategies include selecting quiet locations, eliminating potential distractions and supplementing communication with nonverbal cues. It is important to recognize that the effectiveness of these strategies may vary depending on the individual and the specific communication context.

The association between longitudinal declines in speech sound accuracy and speech intelligibility in speakers with amyotrophic lateral sclerosis

ABSTRACT The purpose of this study was to examine how neurodegeneration secondary to amyotrophic lateral sclerosis (ALS) impacts speech sound accuracy over time and how speech sound accuracy, in turn, is related to speech intelligibility. Twenty-one participants with ALS read the Bamboo Passage over multiple data collection sessions across several months. Phonemic and orthographic transcriptions were completed for all speech samples. The percentage of phonemes accurately produced was calculated across each phoneme, sound class (i.e. consonants versus vowels), and distinctive feature (i.e. features involved in Manner of Articulation, Place of Articulation, Laryngeal Voicing, Tongue Height, and Tongue Advancement). Intelligibility was determined by calculating the percentage of words correctly transcribed orthographically by naive listeners. Linear mixed effects models were conducted to assess the decline of each distinctive feature over time and its impact on intelligibility. The results demonstrated that overall phonemic production accuracy had a nonlinear relationship with speech intelligibility and that a subset of features (i.e. those dependent on precise lingual and labial constriction and/or extensive lingual and labial movement) were more important for intelligibility and were more impacted over time than other features. Furthermore, findings revealed that consonants were more strongly associated with intelligibility than vowels, but consonants did not significantly differ from vowels in their decline over time. These findings have the potential to (1) strengthen mechanistic understanding of the physiological constraints imposed by neuronal degeneration on speech production and (2) inform the timing and selection of treatment and assessment targets for individuals with ALS.

Validity of Off-the-Shelf Automatic Speech Recognition for Assessing Speech Intelligibility and Speech Severity in Speakers With Amyotrophic Lateral Sclerosis.

There is increasing interest in using automatic speech recognition (ASR) systems to evaluate impairment severity or speech intelligibility in speakers with dysarthria. We assessed the clinical validity of one currently available off-the-shelf (OTS) ASR system (i.e., a Google Cloud ASR API) for indexing sentence-level speech intelligibility and impairment severity in individuals with amyotrophic lateral sclerosis (ALS), and we provided guidance for potential users of such systems in research and clinic. Using speech samples collected from 52 individuals with ALS and 20 healthy control speakers, we compared word recognition rate (WRR) from the commercially available Google Cloud ASR API (Machine WRR) to clinician-provided judgments of impairment severity, as well as sentence intelligibility (Human WRR). We assessed the internal reliability of Machine and Human WRR by comparing the standard deviation of WRR across sentences to the minimally detectable change (MDC), a clinical benchmark that indicates whether results are within measurement error. We also evaluated Machine and Human WRR diagnostic accuracy for classifying speakers into clinically established categories. Human WRR achieved better accuracy than Machine WRR when indexing speech severity, and, although related, Human and Machine WRR were not strongly correlated. When the speech signal was mixed with noise (noise-augmented ASR) to reduce a ceiling effect, Machine WRR performance improved. Internal reliability metrics were worse for Machine than Human WRR, particularly for typical and mildly impaired severity groups, although sentence length significantly impacted both Machine and Human WRRs. Results indicated that the OTS ASR system was inadequate for early detection of speech impairment and grading overall speech severity. While Machine and Human WRR were correlated, ASR should not be used as a one-to-one proxy for transcription speech intelligibility or clinician severity ratings. Overall, findings suggested that the tested OTS ASR system, Google Cloud ASR, has limited utility for grading clinical speech impairment in speakers with ALS.

Minimally Detectable Change of Speech Intelligibility in Speakers With Multiple Sclerosis and Parkinson's Disease.

This study sought to determine the minimally detectable change (MDC) of sentence intelligibility for speakers with multiple sclerosis (MS) and Parkinson's disease (PD). Speakers included 78 participants consisting of 32 neurologically healthy control speakers, 30 speakers with MS, and 16 speakers with PD. All speakers were recorded reading 11 sentences comprising the Speech Intelligibility Test (SIT), which were subsequently transcribed by inexperienced listeners. Percent correct scores were calculated for each sentence. An average percent correct score was also calculated for each speaker. The MDC at the 95% confidence interval was calculated using the following formula: MDC95 = 1.96 × √2 × standard error of measurement. Speakers were divided into operationally defined categories of severity, and MDCs were calculated for each category to permit comparison to MDCs reported for speakers with amyotrophic lateral sclerosis (ALS). Kruskal-Wallis tests were conducted to compare MDCs between groups and severity categories. The average MDC95 for control speakers was 5.53% (range = 3.21%-7.47%) and was statistically smaller than MDCs for speakers with MS (average = 10.08%, range = 5.30%-15.62%) and those for speakers with PD (average = 10.98%, range = 8.60%-13.98%). Statistical analyses further revealed significant differences between MDCs across severity categories. In agreement with previous work in ALS conducted under similar conditions (i.e., orthographic transcription of SIT sentences in a quiet listening environment), the MDC95 of speech intelligibility ranged from 3% to 10% for speakers with MS and PD who have mildly impaired speech. These estimates are a step toward the development of a universal language with which to evaluate speech changes in a variety of patient populations.

Agreements between speech language pathologists and naïve listeners’ judgements of intelligibility in children with cleft palate

ABSTRACT Speech-language pathologists (SLPs) working in interdisciplinary craniofacial teams need to know how much their judgments of speech intelligibility could predict the patient’s communication difficulties with non-clinical communicative partners. This study examines the relationship between experienced SLPs and naïve listeners in judgments of speech intelligibility in speakers with cleft palate. A speech perception study was conducted using speech samples from 20 speakers with velopharyngeal insufficiency (VPI) following cleft palate. Speech samples were presented to 70 naïve listeners and 10 SLPs who were experts in cleft palate treatment. Speech intelligibility scores from naïve listeners’ orthographic transcriptions were obtained as the percentage of correctly identified words. Speech intelligibility scores from SLPs were obtained using a five-point rating scale. Spearman rank correlation indicated a very high level of overall agreement between naïve listeners and SLPs at the speaker level scores (rs = −.94, p < .001). While the listeners’ judgment seems highly related across highly intelligible speakers, the differences in agreements increase when the speaker is unintelligible. The high correlations between scores for naïve listeners and SLPs suggest that speech intelligibility in children with VPI could be predicted by ratings done in the clinic by expert SLPs.

Effect of Level of Presentation to Listeners on Scaled Speech Intelligibility of Speakers with Dysarthria

Objective: This study examined the effect of intensity level of presentation on scaling of speech intelligibility in speakers with and without dysarthria. Patients and Methods: A total of 50 utterances produced by speakers with dysarthria and healthy speakers were played to 60 listeners in four conditions, which consisted of two different presentation levels (‘high’ vs. ‘low’) and equalization of levels across utterances (‘adjusted’ vs. ‘unadjusted’). Speech intelligibility was scaled by using a direct magnitude estimation technique with and without modulus. Results: A significant decrease in speech intelligibility was indicated when the stimuli were adjusted to have fixed intensity on the most intense vocalic nuclei of each word, while no significant change was found between ‘high’ and ‘low’ presentation level conditions. Conclusion: The findings suggest that an increase in presentation level alone does not result in significant improvement in speech intelligibility ratings. The results are discussed by considering clinical implications in conducting speech therapy with emphasis on intensity variation.

An Acoustic Study of the Relationships Among Neurologic Disease, Dysarthria Type, and Severity of Dysarthria

This study examined acoustic predictors of speech intelligibility in speakers with several types of dysarthria secondary to different diseases and conducted classification analysis solely by acoustic measures according to 3 variables (disease, speech severity, and dysarthria type). Speech recordings from 107 speakers with dysarthria due to Parkinson's disease, stroke, traumatic brain injury, and multiple system atrophy were used for acoustic analysis and for perceptual judgment of speech intelligibility. Acoustic analysis included 8 segmental/suprasegmental features: 2nd formant frequency slope, articulation rate, voiceless interval duration, 1st moment analysis for fricatives, vowel space, F0, intensity range, and Pairwise Variability Index. The results showed that (a) acoustic predictors of speech intelligibility differed slightly across diseases and (b) classification accuracy by dysarthria type was typically worse than by disease type or severity. These findings were discussed with respect to (a) the relationship between acoustic characteristics and speech intelligibility and (b) dysarthria classification.

Prediction of vowel intelligibility based on acoustic characteristics in speakers with Parkinson’s disease

This study investigated the contribution of acoustic features produced by three vowel contrasts (front–back, high–low, tense–lax) to the intelligibility of the contrast and overall speech intelligibility in speakers with Parkinson’s disease. Multiple regression analyses were used to determine which acoustic measures or combination of measures accounted significantly for the variance in the intelligibility measures. The predictor variables in the multivariate analysis included acoustic measures thought to represent essential perceptual cues to signal the vowel contrast. These findings are discussed in relation to specific areas of production deficiency that are consistent with Parkinson’s disease. [Work supported by NIH R03 DC005902.]

Respiratory dynamics and speech intelligibility in speakers with generalized dystonia

This study investigated aspects of respiratory function, during quiet breathing and monologue, in six adult dystonic subjects and compared the findings to a control group of four neurologically intact adults. Additionally, breathing dynamics were compared with speech intelligibility. Respiratory inductive plethysmography was used to assess breathing rate, periodicity of the breathing pattern, and inspiratory lung volume. Ear oximetry was used to assess arterial blood oxygen saturation. Speech intelligibility was rated by a panel of five judges. Breathing patterns differed between groups; the dystonic subjects showed a faster breathing rate, less rhythmic breathing pattern, decreased lung volume, and apnealike periods accompanied by a decrease in arterial blood oxygen saturation. These differences were observed during quiet breathing and monologue. Decreased speech intelligibility was strongly related to differences in breathing dynamics.