Plug-and-play microphones for recording speech and voice with smart devices

INTRODUCTION Smart devices are widely available and capable of quickly recording and uploading speech segments for health-related analysis. The switch from laboratory recordings with professional-grade microphone set ups to remote, smart device-based recordings offers immense potential for the scalability of voice assessment. Yet, a growing body of literature points to a wide heterogeneity among acoustic metrics for their robustness to variation in recording devices. The addition of consumer-grade plug-and-play microphones has been proposed as a possible solution. Our aim was to assess if the addition of consumer-grade plug-and-play microphones increase the acoustic measurement agreement between ultra-portable devices and a reference microphone. METHODS Speech was simultaneously recorded by a reference high-quality microphone commonly used in research, and by two configurations with plug-and-play microphones. Twelve speech-acoustic features were calculated using recordings from each microphone to determine the agreement intervals in measurements between microphones. Agreement intervals were then compared to expected deviations in speech in various neurological conditions. Each microphone's response to speech and to silence were characterized through acoustic analysis to explore possible reasons for differences in acoustic measurements between microphones. The statistical differentiation of two groups, neurotypical and people with Multiple Sclerosis, using metrics from each tested microphone was compared to that of the reference microphone. RESULTS The two consumer-grade plug-and-play microphones favoured high frequencies (mean centre of gravity difference ≥ +175.3Hz) and recorded more noise (mean difference in signal-to-noise ≤ -4.2dB) when compared to the reference microphone. Between consumer-grade microphones, differences in relative noise were closely related to distance between the microphone and the speaker's mouth. Agreement intervals between the reference and consumer-grade microphones remained under disease-expected deviations only for fundamental frequency (f0, agreement interval ≤0.06Hz), f0 instability (f0 CoV, agreement interval ≤0.05%) and for tracking of second formant movement (agreement interval ≤1.4Hz/millisecond). Agreement between microphones was poor for other metrics, particularly for fine timing metrics (mean pause length and pause length variability for various tasks). The statistical difference between the two groups of speakers was smaller with the plug-and-play than with the reference microphone. CONCLUSION Measurement of f0 and F2 slope were robust to variation in recording equipment while other acoustic metrics were not. Thus, the tested plug-and-play microphones should not be used interchangeably with professional-grade microphones for speech analysis. Plug-and-play microphones may assist in equipment standardization within speech studies, including remote or self-recording, possibly with small loss in accuracy and statistical power as observed in this study.

Disease Delineation for Multiple Sclerosis, Friedreich Ataxia, and Healthy Controls Using Supervised Machine Learning on Speech Acoustics

Neurodegenerative disease often affects speech. Speech acoustics can be used as objective clinical markers of pathology. Previous investigations of pathological speech have primarily compared controls with one specific condition and excluded comorbidities. We broaden the utility of speech markers by examining how multiple acoustic features can delineate diseases. We used supervised machine learning with gradient boosting (CatBoost) to delineate healthy speech from speech of people with multiple sclerosis or Friedreich ataxia. Participants performed a diadochokinetic task where they repeated alternating syllables. We subjected 74 spectral and temporal prosodic features from the speech recordings to machine learning. Results showed that Friedreich ataxia, multiple sclerosis and healthy controls were all identified with high accuracy (over 82%). Twenty-one acoustic features were strong markers of neurodegenerative diseases, falling under the categories of spectral qualia, spectral power, and speech rate. We demonstrated that speech markers can delineate neurodegenerative diseases and distinguish healthy speech from pathological speech with high accuracy. Findings emphasize the importance of examining speech outcomes when assessing indicators of neurodegenerative disease. We propose large-scale initiatives to broaden the scope for differentiating other neurological diseases and affective disorders.

Acoustic Speech Analytics Are Predictive of Cerebellar Dysfunction in Multiple Sclerosis

Speech production relies on motor control and cognitive processing and is linked to cerebellar function. In diseases where the cerebellum is impaired, such as multiple sclerosis (MS), speech abnormalities are common and can be detected by instrumental assessments. However, the potential of speech assessments to be used to monitor cerebellar impairment in MS remains unexplored. The aim of this study is to build an objectively measured speech score that reflects cerebellar function, pathology and quality of life in MS. Eighty-five people with MS and 21 controls participated in the study. Speech was independently assessed through objective acoustic analysis and blind expert listener ratings. Cerebellar function and overall disease disability were measured through validated clinical scores; cerebellar pathology was assessed via magnetic resonance imaging, and validated questionnaires informed quality of life. Selected speech variables were entered in a regression model to predict cerebellar function. The resulting model was condensed into one composite speech score and tested for prediction of abnormal 9-hole peg test (9HPT), and for correlations with the remaining cerebellar scores, imaging measurements and self-assessed quality of life. Slow rate of syllable repetition and increased free speech pause percentage were the strongest predictors of cerebellar impairment, complemented by phonatory instability. Those variables formed the acoustic composite score that accounted for 54% of variation in cerebellar function, correlated with cerebellar white matter volume (r = 0.3, p = 0.017), quality of life (r = 0.5, p < 0.001) and predicted an abnormal 9HPT with 85% accuracy. An objective multi-feature speech metric was highly representative of motor cerebellar impairment in MS.

Speech discrimination performance in multiple sclerosis dataset

The most complex interactions between human beings occur through speech, and often in the presence of background noise. Understanding speech in noisy environments requires the integrity of highly integrated and widespread auditory networks likely to be impacted by multiple sclerosis (MS) related neurogenic injury. Despite the impact auditory communication has on a person's ability to navigate the world, build relationships, and maintain employability; studies of speech-in-noise (SiN) perception in people with MS (pwMS) have been minimal to date. Thus, this paper presents a dataset related to the acquisition of pure-tone thresholds, SiN performance and questionnaire responses in age-matched controls and pwMS. Bilateral pure-tone hearing thresholds were obtained at frequencies of 250 hertz (Hz), 500 Hz, 750 Hz, 1000 Hz, 1500 Hz, 2000 Hz, 4000 Hz, 6000 Hz and 8000 Hz, and hearing thresholds were defined as the lowest level at which the tone was perceived 50% of the time. Thresholds at 500 Hz, 1000 Hz, 2000 Hz and 4000 Hz were used to calculate the four-tone average for each participant, and only those with a bilateral four tone average of ≤ 25 dB HL were included in the analysis. To investigate SiN performance in pwMS, pre-recorded Bamford-Kowal-Bench (BKB) sentences were presented binaurally through headphones at five signal-to-noise ratios (SNR) in two noise conditions: speech-weighted noise and multi-talker babble. Participants were required to verbally repeat each sentence they had just heard; or indicate their inability to do so. A 33-item questionnaire, based on validated inventories for specific adult clinical populations with abnormal auditory processing, was used to evaluate auditory processing in daily life for pwMS. For analysis, pwMS were grouped according to their Expanded Disability Status Scale (EDSS) score as rated by a neurologist. PwMS with EDSS scores ≤ 1.5 were classified as ‘mild’ (n = 20); between 2 and 4.5 as ‘moderate’ (n = 16) and between 5 and 7 as ‘advanced’ (n = 10) and were compared to neurologically healthy controls (n = 38). The outcomes of the SiN task conducted in pwMS can be found in Iva et al., (2021). The present data has important implications for the timing and delivery of preparatory education to patients, family, and caregivers about communication abilities in pwMS. This dataset will also be valuable for the reuse/reanalysis required for future investigations into the clinical utility of SiN tasks to monitor disease progression.

Effects of face masks on acoustic analysis and speech perception: Implications for peri-pandemic protocols

Wearing face masks (alongside physical distancing) provides some protection against infection from COVID-19. Face masks can also change how people communicate and subsequently affect speech signal quality. This study investigated how three common face mask types (N95, surgical, and cloth) affected acoustic analysis of speech and perceived intelligibility in healthy subjects. Acoustic measures of timing, frequency, perturbation, and power spectral density were measured. Speech intelligibility and word and sentence accuracy were also examined using the Assessment of Intelligibility of Dysarthric Speech. Mask type impacted the power distribution in frequencies above 3 kHz for the N95 mask, and above 5 kHz in surgical and cloth masks. Measures of timing and spectral tilt mainly differed with N95 mask use. Cepstral and harmonics to noise ratios remained unchanged across mask type. No differences were observed across conditions for word or sentence intelligibility measures; however, accuracy of word and sentence translations were affected by all masks. Data presented in this study show that face masks change the speech signal, but some specific acoustic features remain largely unaffected (e.g., measures of voice quality) irrespective of mask type. Outcomes have bearing on how future speech studies are run when personal protective equipment is worn.

Speech discrimination impairments as a marker of disease severity in multiple sclerosis

BACKGROUND

Multiple Sclerosis (MS) pathology is likely to disrupt central auditory pathways, thereby affecting an individual's ability to discriminate speech from noise. Despite the importance of speech discrimination in daily communication, it's characterization in the context of MS remains limited. This cross-sectional study evaluated speech discrimination in MS under "real world" conditions where sentences were presented in ecologically valid multi-talker speech or broadband noise at several signal-to-noise ratios (SNRs).

METHODS

Pre-recorded Bamford-Kowal-Bench sentences were presented at five signal-to-noise ratios (SNR) in one of two background noises: speech-weighted noise and eight-talker babble. All auditory stimuli were presented via headphones to control (n = 38) and MS listeners with mild (n = 20), moderate (n = 16) and advanced (n = 10) disability. Disability was quantified by the Kurtzke Expanded Disability Status Scale (EDSS) and scored by a neurologist. All participants passed a routine audiometric examination.

RESULTS

Despite normal hearing, MS psychometric discrimination curves which model the relationship between signal-to-noise ratio (SNR) and sentence discrimination accuracy in speech-weighted noise and babble did not change in slope (sentences/dB) but shifted to higher SNRs (dB) compared to controls. The magnitude of the shift in the curve systematically increased with greater disability. Furthermore, mixed-effects models identified EDSS score as the most significant predictor of speech discrimination in noise (odds ratio = 0.81; p < 0.001). Neither age, sex, disease phenotype or disease duration were significantly associated with speech discrimination performance in noise. Only MS listeners with advanced disability self-reported audio-attentional difficulty in a questionnaire designed to reflect auditory processing behaviours in daily life.

CONCLUSION

Speech discrimination performance worsened systematically with greater disability, independent of age, sex, education, disease duration or disease phenotype. These results identify novel auditory processing deficits in MS and highlight that speech discrimination tasks may provide a viable non-invasive and sensitive means for disease monitoring in MS.

Speech metrics, general disability, brain imaging and quality of life in multiple sclerosis

BACKGROUND AND PURPOSE

Objective measurement of speech has shown promising results to monitor disease state in multiple sclerosis. In this study, we characterize the relationship between disease severity and speech metrics through perceptual (listener based) and objective acoustic analysis. We further look at deviations of acoustic metrics in people with no perceivable dysarthria.

METHODS

Correlations and regression were calculated between speech measurements and disability scores, brain volume, lesion load and quality of life. Speech measurements were further compared between three subgroups of increasing overall neurological disability: mild (as rated by the Expanded Disability Status Scale ≤2.5), moderate (≥3 and ≤5.5) and severe (≥6).

RESULTS

Clinical speech impairment occurred majorly in people with severe disability. An experimental acoustic composite score differentiated mild from moderate (P < 0.001) and moderate from severe subgroups (P = 0.003), and correlated with overall neurological disability (r = 0.6, P < 0.001), quality of life (r = 0.5, P < 0.001), white matter volume (r = 0.3, P = 0.007) and lesion load (r = 0.3, P = 0.008). Acoustic metrics also correlated with disability scores in people with no perceivable dysarthria.

CONCLUSIONS

Acoustic analysis offers a valuable insight into the development of speech impairment in multiple sclerosis. These results highlight the potential of automated analysis of speech to assist in monitoring disease progression and treatment response.

OnabotulinumtoxinA treatment for MS-tremor modifies fMRI tremor response in central sensory-motor integration areas

BACKGROUND

Treatment of tremor in MS is an unmet need. OnabotulinumtoxinA (BoNT-A) has shown promising results; however, little is known regarding its effects on the brain. The clinical presentation of tremor MS is shown to depend on subcortical neural damage and cortical neural plasticity. This study aimed to identify effects of onabotulinumtoxinA (BoNT-A) on brain activation in MS and upper-limb tremor using functional MRI.

METHODS

Forty-three MS participants with tremor were randomized to receive intramuscular injections of placebo (n = 22) or BoNT-A (n = 21). Tremor was quantified using the Bain score (0-10) for severity, handwriting and Archimedes drawing at baseline, 6 weeks and 12 weeks. Functional MRI activation within two previously identified clusters, ipsilateral inferior parietal cortex (IPL) and premotor/supplementary motor cortex (SMC) of compensatory activity, was measured at baseline and 6 weeks.

RESULTS

Treatment with BoNT-A resulted in improved handwriting tremor at 6 weeks (p = 0.049) and 12 weeks (p = 0.014), and tremor severity -0.79 (p = 0.007) at 12 weeks. Furthermore, the patients that received BoNT-A showed a reduction in activation within the IPL (p = 0.034), but not in the SMC. The change in IPL activation correlated with the reduction in tremor severity from baseline to 12 weeks (β = 0.608; p = 0.015) in the BoNT-A group. No tremor and fMRI changes were seen in the placebo treated group.

CONCLUSION

We have shown that reduction in MS-tremor severity after intramuscular injection with BoNT-A is associated with changes in brain activity in sensorimotor integration regions.

Functional neuroplasticity in response to cerebello-thalamic injury underpins the clinical presentation of tremor in multiple sclerosis

BACKGROUND

Tremor is present in almost half of multiple sclerosis (MS) patients. The lack of understanding of its pathophysiology is hampering progress in development of treatments.

OBJECTIVES

To clarify the structural and functional brain changes associated with the clinical phenotype of upper limb tremor in people with MS.

METHODS

Fifteen healthy controls (46.1 ± 15.4 years), 27 MS participants without tremor (46.7 ± 11.6 years) and 42 with tremor (46.6 ± 11.5 years) were included. Tremor was quantified using the Bain score (0-10) for overall severity, handwriting and Archimedes spiral drawing. Functional magnetic resonance imaging activations were compared between participants groups during performance of a joystick task designed to isolate tremulous movement. Inflammation and atrophy of cerebello-thalamo-cortical brain structures were quantified.

RESULTS

Tremor participants were found to have atrophy of the cerebellum and thalamus, and higher ipsilateral cerebellar lesion load compared to participants without tremor (p < 0.020). We found higher ipsilateral activation in the inferior parietal lobule, the premotor cortex and supplementary motor area in MS tremor participants compared to MS participants without tremor during the joystick task. Finally, stronger activation in those areas was associated with lower tremor severity.

CONCLUSION

Subcortical neurodegeneration and inflammation along the cerebello-thalamo-cortical and cortical functional neuroplasticity contribute to the severity of tremor in MS.

Validation of a precision tremor measurement system for multiple sclerosis

BACKGROUND

Tremor is a debilitating symptom of Multiple Sclerosis (MS). Little is known about its pathophysiology and treatments are limited. Clinical trials investigating new interventions often rely on subjective clinical rating scales to provide supporting evidence of efficacy.

NEW METHOD

We present a novel instrument (TREMBAL) which uses electromagnetic motion capture technology to quantify MS tremor. We aim to validate TREMBAL by comparison to clinical ratings using regression modelling with 310 samples of tremor captured from 13 MS participants who performed five different hand exercises during several follow-up visits. Minimum detectable change (MDC) and test-retest reliability were calculated and comparisons were made between MS tremor and data from 12 healthy volunteers.

RESULTS

Velocity of the index finger was most congruent with clinical observation. Regression modelling combining different features, sensor configurations, and labelling exercises did not improve results. TREMBAL MDC was 84% of its initial measurement compared to 91% for the clinical rating. Intra-class correlations for test-retest reliability were 0.781 for TREMBAL and 0.703 for clinical ratings. Tremor was lower (p =  0.002) in healthy subjects.

COMPARISON WITH EXISTING METHODS

Subjective scales have low sensitivity, suffer from ceiling effects, and mitigation against inter-rater variability is challenging. Inertial sensors are ubiquitous, however, their output is nonlinearly related to tremor frequency, compensation is required for gravitational artefacts, and their raw data cannot be intuitively comprehended.

CONCLUSIONS

TREMBAL, compared with clinical ratings, gave measures in agreement with clinical observation, had marginally lower MDC, and similar test-retest reliability.

Novel Functional MRI Task for Studying the Neural Correlates of Upper Limb Tremor

Introduction: Tremor of the upper limbs is a disabling symptom that is present during several neurological disorders and is currently without treatment. Functional MRI (fMRI) is an essential tool to investigate the pathophysiology of tremor and aid the development of treatment options. However, no adequately or standardized protocols for fMRI exists at present. Here we present a novel, online available fMRI task that could be used to assess the in vivo pathology of tremor.

Objective: This study aims to validate the tremor-evoking potential of the fMRI task in a small group of tremor patients outside the scanner and assess the reproducibility of the fMRI task related activation in healthy controls.

Methods: Twelve HCs were scanned at two time points (baseline and after 6-weeks). There were two runs of multi-band fMRI and the tasks included a “brick-breaker” joystick game. The game consisted of three conditions designed to control for most of the activation related to performing the task by contrasting the conditions: WATCH (look at the game without moving joystick), MOVE (rhythmic left/right movement of joystick without game), and PLAY (playing the game). Task fMRI was analyzed using FSL FEAT to determine clusters of activation during the different conditions. Maximum activation within the clusters was used to assess the ability to control for task related activation and reproducibility. Four tremor patients have been included to test ecological and construct validity of the joystick task by assessing tremor frequencies captured by the joystick.

Results: In HCs the game activated areas corresponding to motor, attention and visual areas. Most areas of activation by our game showed moderate to good reproducibility (intraclass correlation coefficient (ICC) 0.531–0.906) with only inferior parietal lobe activation showing poor reproducibility (ICC 0.446). Furthermore, the joystick captured significantly more tremulous movement in tremor patients compared to HCs (p = 0.01) during PLAY, but not during MOVE.

Conclusion: Validation of our novel task confirmed tremor-evoking potential and reproducibility analyses yielded acceptable results to continue further investigations into the pathophysiology of tremor. The use of this technique in studies with tremor patient will no doubt provide significant insights into the treatment options.