Speech Recognition Research Articles

Benefits of speech recognition in noise using remote microphones for people with typical hearing

IntroductionRemote microphone (RM) systems are designed to enhance speech recognition in noisy environments by improving the signal-to-noise ratio (SNR) for individuals with typical hearing (TH) and hearing impairment (HI). The aim of this investigation was to evaluate the advantages of speech recognition in noise for individuals with TH in a simulated group setting using two different remote microphones. MethodsA quasi-experimental, repeated-measures design was employed, involving ten participants with TH, ages 20 to 63 years. Each were fit with Roger Focus receivers bilaterally to listen to three RM conditions: Roger Select, Roger Pen, and no technology. Participants were instructed to transcribe sentences that were presented randomly at varying signal-to-noise ratios (SNRs: 0, -5, and -10 dB) from five speakers positioned equidistant around a circular table to simulate a group dining scenario. ResultsSignificant main effects of the technology condition and noise level (p < .05) were found. Participants exhibited superior performance with Roger Select compared to Roger Pen. As expected, recognition rates decreased with lower SNRs across all three technology conditions. ConclusionsTo enhance speech recognition in group settings for individuals with TH, the utilization of the Roger Select microphone in conjunction with bilateral Roger Focus receivers is recommended over the Roger Pen

EXPRESS: Prior multisensory learning can facilitate auditory-only voice-identity and speech recognition in noise.

Seeing the visual articulatory movements of a speaker, while hearing their voice, helps with understanding what is said. This multisensory enhancement is particularly evident in noisy listening conditions. Multisensory enhancement also occurs even in auditory-only conditions: auditory-only speech and voice-identity recognition is superior for speakers previously learned with their face, compared to control learning; an effect termed the "face-benefit". Whether the face-benefit can assist in maintaining robust perception in increasingly noisy listening conditions, similar to concurrent multisensory input, is unknown. Here, in two behavioural experiments, we examined this hypothesis. In each experiment, participants learned a series of speakers' voices together with their dynamic face, or control image. Following learning, participants listened to auditory-only sentences spoken by the same speakers and recognised the content of the sentences (speech recognition, Experiment 1) or the voice-identity of the speaker (Experiment 2) in increasing levels of auditory noise. For speech recognition, we observed that 14/30 participants (47%) showed a face-benefit. While 19/25 participants (76%) showed a face-benefit for voice-identity recognition. For those participants who demonstrated a face-benefit, the face-benefit increased with auditory noise levels. Taken together, the results support an audio-visual model of auditory communication and suggest that the brain can develop a flexible system in which learned facial characteristics are used to deal with varying auditory uncertainty.

Human-machine Voice and Human-machine Competition: Practical Landscape, Pattern Changes and Industrial Challenges of Generative AI-Enabled Audiobook Publishing

At present, generative artificial intelligence has a profound impact on the audiobook publishing industry chain, and exerts unprecedented transformative power in all aspects of audiobook publishing. This paper reviews the specific technical forms of generative artificial intelligence used in the process of audiobook publishing, including text-to-speech, speech recognition, intelligent speech interaction, voiceprint recognition, voice anonymization, etc., and shows the practical picture of how these technologies influence audiobook publishing in different stages of content creation, sound presentation, recording and editing, quality control, terminal playback and readers feedback. It analyzes the pattern changes of the audiobook publishing industry brought about by generative artificial intelligence, with "human-machine voice" as the core, and then discusses the challenges and reflections faced by the audiobook publishing industry under the trend of "human-machine competition and cooperation" from the aspects of underlying logic, industry pain points, fundamental principle, reading environment and technical ethics. Therefore, generative artificial intelligence technology is like a double-edged sword, audiobook publishers should fully understand its nature, conform to the trend of technology, and use it to enable new quality productive forces of audiobook publishing industry.

Development and Initial Testing of an Artificial Intelligence-Based Virtual Reality Companion for People Living with Dementia in Long-Term Care.

Background/Objectives: Feelings of loneliness are common in people living with dementia (PLWD) in long-term care (LTC). The goals of this study were to describe the development of a novel virtual companion for PLWD living in LTC and assess its feasibility and acceptability. Methods: The computer-generated virtual companion, presented using a head-mounted virtual reality display, was developed in two stages. In Stage 1, the virtual companion asked questions designed to encourage conversation and reminiscence. In Stage 2, more powerful artificial intelligence tools allowed the virtual companion to engage users in nuanced discussions on any topic. PLWD in LTC tested the application at each stage to assess feasibility and acceptability. Results: Ten PLWD living in LTC participated in Stage 1 (4 men and 6 women; average 82 years old) and Stage 2 (2 men and 8 women; average 87 years old). Session lengths ranged from 0:00 to 5:30 min in Stage 1 and 0:00 to 53:50 min in Stage 2. Speech recognition issues and a limited repertoire of questions limited acceptance in Stage 1. Enhanced conversational ability in Stage 2 led to intimate and meaningful conversations with many participants. Many users found the head-mounted display heavy. There were no complaints of simulator sickness. The virtual companion was best suited to PLWD who could engage in reciprocal conversation. After Stage 2, response latency was identified as an opportunity for improvement in future versions. Conclusions: Virtual reality and artificial intelligence can be used to create a virtual companion that is acceptable and enjoyable to some PLWD living in LTC. Ongoing innovations in hardware and software will allow future iterations to provide more natural conversational interaction and an enhanced social experience.

The effect of percutaneous and transcutaneous BAHA on hearing and subjective auditory abilities: A comparative study.

Hearing loss significantly affects communication, social interactions, and the overall quality of life. The bone-anchored hearing aid (BAHA) is an implantable system that bypasses the outer and middle ear to directly stimulate the cochlea through bone conduction. This study aimed to compare hearing performance and subjective auditory ability improvements between transcutaneous and percutaneous BAHA devices using audiological assessments and Speech, Spatial and Qualities of Hearing Scale. This cross-sectional prospective study included 29 participants aged 14 to 69 years who had used BAHA for at least 6 months. Both Cochlear Baha System's percutaneous (connect) and transcutaneous (attract) implants were evaluated. Audiological assessments involved pure-tone audiometry, speech recognition threshold, and free-field (FF) audiometry, while subjective auditory ability was measured using the Turkish Speech, Spatial and Qualities of Hearing scale (Tr-SSQ). Significant improvements in FF audiometry averages and speech recognition thresholds were observed with BAHA compared to without BAHA (P < .001). Both implant types provided similar FF averages, speech audiometry results, and Tr-SSQ outcomes, with no significant differences between them. Tr-SSQ scores showed substantial satisfaction, indicating significant improvements in speech perception, spatial perception, and hearing quality with BAHA (P < .001). The findings align with previous research, demonstrating that BAHA is a reliable and effective solution for hearing rehabilitation. The study also emphasized the importance of using both audiological test results and daily hearing function scales to comprehensively evaluate the benefits of hearing rehabilitation in real-world environments. In conclusion, BAHA, regardless of the implant type, can provide predictable and lasting improvements in hearing thresholds and daily hearing abilities, making it a valuable option for patients with conductive hearing loss.

Hypernetworks for Personalizing ASR to Atypical Speech

Abstract Parameter-efficient fine-tuning (PEFT) for personalizing automatic speech recognition (ASR) has recently shown promise for adapting general population models to atypical speech. However, these approaches assume a priori knowledge of the atypical speech disorder being adapted for—the diagnosis of which requires expert knowledge that is not always available. Even given this knowledge, data scarcity and high inter-/intra-speaker variability further limit the effectiveness of traditional fine-tuning. To circumvent these challenges, we first identify the minimal set of model parameters required for ASR adaptation. Our analysis of each individual parameter’s effect on adaptation performance allows us to reduce Word Error Rate (WER) by half while adapting 0.03% of all weights. Alleviating the need for cohort-specific models, we next propose the novel use of a meta-learned hypernetwork to generate highly individualized, utterance-level adaptations on-the-fly for a diverse set of atypical speech characteristics. Evaluating adaptation at the global, cohort, and individual-level, we show that hypernetworks generalize better to out-of-distribution speakers, while maintaining an overall relative WER reduction of 75.2% using 0.1% of the full parameter budget.

Financial Fraud Detection and Prevention

Financial fraud has always been a serious challenge in the financial sector. With the continuous development of technology, fraud in the financial market has become increasingly complex and hidden. Therefore, financial fraud detection and prevention have become particularly important to our lives. But as of today, the financial fraud detection methods that have emerged often leave something to be desired. Traditional detection methods based on rules and statistical methods perform poorly when processing large-scale and high-dimensional data, and are prone to false positives and false negatives. Moreover, as fraud techniques continue to evolve, the adaptability of traditional methods is also challenged. As a powerful machine learning technology, deep learning has excellent feature extraction and pattern recognition capabilities, and has achieved remarkable achievements in various fields, including image recognition, natural language processing, and speech recognition. In the financial field, the application of deep learning is also gradually emerging.

Updates on Clinical Language Sampling Practices: A Survey of Speech-Language Pathologists Practicing in the United States.

Language sample analysis (LSA) provides many benefits for assessing, identifying therapy goals, and monitoring the progress of children with language disorders. Despite these widely recognized advantages, previous surveys suggest the declining use of LSA by speech-language pathologists (SLPs). This study aimed to provide updates on clinical LSA use following the recent introduction of two new LSA protocols, namely, the Sampling Utterances and Grammatical Analysis Revised (SUGAR) protocol and the Computerized Language Analysis KIDEVAL program. Survey data from SLPs practicing in the United States (N = 337) were used to examine rates of LSA use, methods, and protocols. Factors predicting LSA use and reported facilitators and barriers were also examined. Results indicated that 60% of SLPs used LSA in the past year. LSA skill level, training, and serving preschool or elementary school children predicted LSA use, whereas workplace, caseload, and years of experience were not significant predictors. Most SLPs reported using self-designed LSA protocols (62%), followed by Systematic Analysis of Language Transcripts (23%) and SUGAR (12%) protocols. SLPs who did not use LSA reported limited time (74%), limited resources (59%), and limited expertise (41%) as barriers and identified additional training on LSA computer programs (52%) and access to automatic speech recognition programs (49%) as facilitators to their adoption of LSA. Reported rates of LSA use and methods were consistent with previous survey findings. This study's findings highlight the ongoing needs for more extensive preprofessional training in LSA.

The role of GDF15 in attenuating noise-induced hidden hearing loss by alleviating oxidative stress

Noise-induced hidden hearing loss (HHL) is a newly uncovered form of hearing impairment that causes hidden damage to the cochlea. Patients with HHL do not have significant abnormalities in their hearing thresholds, but they experience impaired speech recognition in noisy environments. However, the mechanisms underlying HHL remain unclear. In this study, we developed single-cell transcriptome profiles of the cochlea of mice with HHL, detailing changes in individual cell types. Our study revealed a transient threshold shift, reduced auditory brainstem response wave I amplitude, and decreased number of ribbon synapses in HHL mice. Our findings suggest elevated oxidative stress and GDF15 expression in cochlear hair cells of HHL mice. Notably, the upregulation of GDF15 attenuated oxidative stress and auditory impairment in the cochlea of HHL mice. This suggests that a therapeutic strategy targeting GDF15 may be efficacious against HHL.Graphical HHL mice had a transient threshold shift, reduced ABR wave I amplitude, and decreased number of ribbon synapses.HHL mice's cochlear hair cells exhibited increased oxidative stress and elevated GDF15 expression.Upregulation of GDF15 attenuated oxidative stress and auditory damage in the cochlea of HHL mice, implying that GDF15-targeted treatment techniques may be useful for HHL.

Music source feature extraction based on improved attention mechanism and phase feature

Music source feature extraction is an important research direction in music information retrieval and music recommendation system. To extract the features of music sources more effectively, the study introduces the jump attention mechanism and combines it with the convolutional attention module. Also, a feature extraction module based on Unet + + and spatial attention module is proposed. In addition, the phase feature information of the mixed music signals is utilized to improve the network performance. Results showed that this model was studied to perform well in music source separation experiments of vocals and accompaniment. For vocal separation on the MIR-1K dataset, the model achieves 11.25 dB, 17.34 dB, and 13.83 dB for each metric, respectively. Meanwhile, for drum separation on the DSD100 dataset, the model achieves a median signal-to-source distortion ratio of 4.36 dB, which is 2.91 dB better than that of the Spectral Hierarchical Network model. For the separation of the bass sound and the human voice, the model's in the separation of bass and human voice, the median distortion ratio of the model is as high as 4.87 dB and 6.09 dB, which is better than that of the Spectral Hierarchical Network model. This indicates the significant performance advantages in feature extraction and separation of music sources, and it has important application values in music production and speech recognition.

Equivalence and test–retest reliability of Tulu sentence lists for measuring speech recognition threshold in noise among individuals with sensorineural hearing loss

BackgroundThe difficulty in understanding speech becomes worse in the presence of background noise for individuals with sensorineural hearing loss. Speech-in-noise tests help to assess this difficulty. Previously, the Tulu sentence lists have been assessed for their equivalency to measure speech recognition threshold in noise among individuals with normal hearing. The present study aimed to determine the equivalence and test–retest reliability of Tulu sentence lists for measuring speech recognition threshold in noise among individuals with sensorineural hearing loss.ResultsThe SNR-50 was measured for 13 sentence lists in 20 Tulu-speaking individuals with mild to moderate sensorineural hearing loss. Retesting was done by administering all lists to eight participants after an average of 25.25 days (SD = 19.44). Friedman test was administered to check for the list equivalency. Intraclass correlation coefficient was measured to assess test–retest reliability. A regression analysis was performed to understand the influence of pure-tone average on SNR-50. A Kruskal–Wallis test was administered to check the statistical significance of the SNR-50 obtained across different configurations and degrees of hearing loss. Nine of the 13 Tulu sentence lists (lists 2, 4, 5, 6, 9, 10, 11, 12, and 13) were equivalent in individuals with sensorineural hearing loss. The mean SNR-50 for these nine lists was 1.13 dB (SD = 2.04 dB). The test–retest reliability was moderate (ICC = 0.727). The regression analysis showed that a pure-tone average accounted for 24.7% of the variance in SNR-50 data (p = 0.026). Individuals with mild to moderate hearing loss obtained the worst SNR-50, followed by mild and high-frequency hearing loss.ConclusionNine Tulu sentence lists are equivalent and reliable and can be used to measure speech recognition threshold in noise among individuals with sensorineural hearing loss who are Tulu speakers.

An improved wiener filter-based method for identifying stall inception of transonic compressor

The development of the modern aviation industry poses high demands on the design of aircraft engines recently. However, the stability of compressor flow is one of the key factors affecting further improvements in engine performance. The design of next-generation aircraft engines imposes higher requirements on compressor loading, which leads to the emergence of many new stall inceptions. As a result, the onset and evolution of flow instability become more complex. For accurately capturing stall inceptions of transonic compressors, the strong pressure disturbances caused by shock waves at the blade tip and the complex flow within blade passages result in significant challenges. To address this issue, this study draws inspiration from the design methods of Wiener filters in the field of speech recognition. Based on the characteristic signal mutations during rotating stall in compressors, a Wiener filter approach that uses time-delayed signals as noise estimates during filter training is developed. The method can be used for both offline and online analysis. It was applied to analyze the stall signals of a single rotor from a 1.5-stage transonic axial compressor under distorted inlet conditions at transonic rotational speeds and the entire stage under uniform inlet conditions at subsonic rotational speeds. The results indicate that, under inlet distortion, the compressor generates disturbance signals in the distorted sector before stall, and the earliest spike-inception disturbance occurs at the circumferential position of the rotor leaving the distorted sector. Under uniform inlet conditions, random disturbances could be detected throughout the circumference before stall onset, developing into spike waves at a circumferential location that subsequently triggered stall. Compared to conventional low-pass filters, discrete wavelet transforms, and empirical mode decomposition, the Wiener filter yielded more prominent spike wave structures in the filtered signals. Under distorted inlet conditions, the Wiener-filtered signals showed a 1 % decrease in autocorrelation coefficient and a 3.7 % increase in root mean square (RMS) upon the appearance of spike waves, more pronounced than the 0.5 % decrease in autocorrelation coefficient and 1.8 % increase in RMS achieved by conventional methods. Under uniform inlet conditions, the Wiener filter also detected a 9.7 % increase in RMS upon the appearance of spike waves, more pronounced than the 6.3 % increase observed with conventional methods.

Automatic subtitles increase accuracy and decrease cognitive load in simultaneous interpreting

Abstract This study examines the effect of real-time subtitles generated by automatic speech recognition (ASR) technology on interpreting accuracy and interpreters’ cognitive load. Multiple measurements — including interpreting accuracy, the NASA-TLX for subjective ratings of cognitive load, eye-tracking and theta power as indicated by EEG recordings — were applied. Twenty-three professional simultaneous interpreters worked with a video recording of a speech presented in five conditions: a baseline without subtitles and then with subtitles of varying levels of precision (100%, 95%, 90% and 80%). The results reveal that the presence of subtitles significantly improved interpreting accuracy, with a suggested optimal precision rate of 90% or higher. The interpreters looked more at the subtitles, regardless of their level of precision, than the speaker. Contrary to our predictions, the presence of subtitles decreased, rather than increased, the cognitive load (although this outcome was shown by the EEG data only and not by the self-reported data). We conclude that the cognitive cost of processing subtitles as an additional information channel is offset by the cognitive gain achieved through visual prompting. The study highlights a complex effect of subtitles on interpreting, with such factors as subtitle presence and precision modulating the interpreters’ cognitive load in such a workflow.

Augmented dialectal speech recognition for AI-based neuropsychological scale assessment in Alzheimer’s disease

Alzheimer’s disease (AD) is a prevalent and widespread neurodegenerative disorder among the older adult population worldwide. Among the numerous cognitive screening methods, neuropsychological scale assessments (NSA) are the widely utilized screening tools in clinical practice. The NSA places significant emphasis on speech-related questions, and thus the role of Automatic Speech Recognition (ASR) technology in AI-based NSA evaluations becomes particularly critical. However, the majority of ASR research pays limited attention to the study of speech recognition for various dialects. One of the primary reasons and main challenges for the scarcity of research in dialectal speech recognition is the limited availability of annotated dialectal data. Furthermore, the older adults’ unclear pronunciation leads to erroneous recognition of similar phonetic words by ASR. To overcome the challenges of limited annotated dialect data in various regions and the issues related to confusable pronunciation of older adults, we propose a dialectal collaborative encoder mechanism and a machine reading comprehension augmented model (MRCAM) that automatically corrects the texts in the NSA. The dialectal collaborative encoder mechanism utilizes layer frozenness to transfer recognition ability to the dialectal model. Meanwhile, the MRCAM model fills the gap by combining speech recognition algorithms with natural language processing (NLP) technology to recognize the dialects spoken by older adults. The experimental evaluation results indicate that the proposed augmented model achieves over 90% consistency with the physician’s judgment, providing the premise of the proposed methods for the task of automated screening for AD based on ASR.

Comparative study on the accuracy of speech recognition using a contact microphone attached to the surface of the head and neck

The accuracy of speech recognition through an air-conducted microphone can be less accurate under a highly noisy environment or when the volume of the user’s voice is relatively low. One solution to this problem is the use of contact microphones. However, neither the microphone locations that provide optimal speech recognition accuracy for each user nor the mechanisms underlying these contact forces have been clarified. In this study, we experimentally investigated the effects of placement, contact force, user gender, and speech recognition platform on the accuracy of speech recognition with contact microphones placed on the surface of the head and neck. The experimental results indicated that the mechanism underlying the influence of each factor on speech recognition accuracy differs for speech acquired at the neck and head locations. In particular, the effect of the user’s gender was significant for the neck-acquired sound, but not the head-acquired sound. The results also revealed that the microphone contact force did not affect the recognition accuracy or user discomfort for the head-acquired sound. Moreover, the results of speech recognition experiments in a simulated noisy environment showed that bone-conducted sounds acquired on the head and neck surfaces were more robust than air-conducted sounds.

Evaluating cochlear implant outcomes in DFNA9 subjects: a comprehensive study on cerebral white matter lesions and vestibular abnormalities.

This study assessed whether the Fazekas score could account for the variability in cochlear implantation (CI) outcomes among individuals with DFNA9 and evaluated signal loss in the semicircular canals (SCCs) on magnetic resonance imaging (MRI) among individuals with DFNA9. This retrospective cross-sectional study included CI recipients with DFNA9. Pre-implantation MRI-scans were reviewed to determine the Fazekas score, localizing and grading cerebral white matter lesions (WML), and identify abnormalities in the SCCs. CI performance was assessed by evaluating phoneme scores one year post-implantation. The function of the SCCs was evaluated using rotatory chair testing with electronystagmography (ENG) and the video Head Impulse Test (vHIT). Forty-five subjects (49 ears) were enrolled. The phoneme scores significantly improved from 35% (IQR 11-50) pre-implantation to 84% (IQR 76-90) one year post-implantation. No correlation was observed between the Fazekas score and the one-year post-implantation phoneme score (rsp=0.003, p = 0.986). Signal loss in at least one SCCs was detected in 97.7% of subjects and 77.8% of ears. There was no correlation between vestibular test results and fluid signal loss in the SCCs on MRI. Most individuals with DFNA9 show improved speech recognition with CI. The observed variability in CI outcomes was not linked to the Fazekas score. Additionally, our study confirms a high prevalence of focal sclerosis in DFNA9. Recognizing the limitations of this study, further research is needed to explore the predictive role of the Fazekas score on CI outcomes and its relationship with vestibular function.

Artificial basilar membrane/hair cell integrated acoustic system for keyword spotting in noisy environments inspired by human cochlea

We report a novel speech recognition method using a noise-robust acoustic sensor system that integrates a spatially frequency-separating sensor with a nonlinear amplification algorithm, mimicking the cochlea’s basilar membrane and hair cells. The multichannel piezoelectric artificial basilar membrane (ABM) sensor detects specific sound frequencies with high sensitivity over 0.2–6 kHz. The signal processing model of the Artificial Hair Cell inspired by the signal transduction mechanism of inner hair cells, simultaneously enhances the frequency selectivity of ABM sensors and improves noise robustness. In a 0 dB SNR noisy environment, it effectively detected the voice signal with a maximum SNR of 57 dB. Furthermore, we converted the frequency-separated signals for speech sounds in various noisy environments into heatmap images and utilized them as input for a CNN-based speech recognition algorithm. Consequently, our system demonstrated noise-robust recognition performance with 94 % accuracy, even in noisy environments.

Collaborative robot acting as scrub nurse for cataract surgery (CRASCS).

Worldwide, healthcare systems are struggling to tackle a nursing shortage. As per the data released by the American Hospital Association, the healthcare workforce could face a loss of around 5 lakh nurses by the end of the year. Consequently, it could result in those deficiencies, which will be 1.1 million instead of 0.6 million. The current nursing scenario in India as per the Indian Nursing Council (INC), which is a board under the ministry and is responsible for legally confirming and maintaining universal standardized training for nursing, is that the 1.96 nurses out of every 1000 Indians that are there are much behind the World Health Organization's (WHO) recommended figure of 3 nurses per 1000. To mitigate the nurse shortage, a collaborative robotic system was designed that can assist with surgical procedures with a collaborative robot acting as a scrub nurse for cataract surgery (CRASCS) represented in Fig. 1. Accordingly, the model has been built to empower a customized 3d printed 5-Degree of freedom robotic arm by tracking the phase of surgery in real-time and automatically supplying the clinician with the ideal equipment that is needed for the particular phase of surgery. The system is supported with one more model which can identify where the surgical equipment is located within the arm range. The system is also supported with voice commands which help in picking up the right surgical equipment in the middle of any phase of the surgery. In this way, the system could be able to potentially handle the shortage of surgical nurses around the world and benefit humanity.

Accuracy of Speech Sound Analysis: Comparison of an Automatic Artificial Intelligence Algorithm With Clinician Assessment.

Automatic speech analysis (ASA) and automatic speech recognition systems are increasingly being used in the treatment of speech sound disorders (SSDs). When utilized as a home practice tool or in the absence of the clinician, the ASA system has the potential to facilitate treatment gains. However, the feedback accuracy of such systems varies, a factor that may impact these gains. The current research analyzes the feedback accuracy of a novel ASA algorithm (Amplio Learning Technologies), in comparison to clinician judgments. A total of 3,584 consonant stimuli, produced by 395 American English-speaking children and adolescents with SSDs (age range: 4-18 years), were analyzed with respect to automatic classification of the ASA algorithm, clinician-ASA agreement, and interclinician agreement. Further analysis of results as related to phoneme acquisition categories (early-, middle-, and late-acquired phonemes) was conducted. Agreement between clinicians and ASA classification for sounds produced accurately was above 80% for all phonemes, with some variation based on phoneme acquisition category (early, middle, late). This variation was also noted for ASA classification into "acceptable," "unacceptable," and "unknown" (which means no determination of phoneme accuracy) categories, as well as interclinician agreement. Clinician-ASA agreement was reduced for misarticulated sounds. The initial findings of Amplio's novel algorithm are promising for its potential use within the context of home practice, as it demonstrates high feedback accuracy for correctly produced sounds. Furthermore, complexity of sound influences consistency of perception, both by clinicians and by automated platforms, indicating variable performance of the ASA algorithm across phonemes. Taken together, the ASA algorithm may be effective in facilitating speech sound practice for children with SSDs, even in the absence of the clinician.

Spontaneous neural activity underlying neutral and happy speech recognition in noise and its association with psychiatric symptoms in patients with schizophrenia

BackgroundDeficits in speech and emotion perception are intertwined with psychiatric symptoms. How the happy prosody embedded in speech affects target speech-in-noise recognition (TSR) and relates to psychiatric symptoms in patients with schizophrenia (SCHs) remains unclear. This study examined spontaneous brain activity underlying happy TSR and its association with psychiatric symptom dimensions in SCHs. MethodsFifty-four SCHs and 59 healthy control participants (HCs) underwent the TSR task, Positive and Negative Syndrome Scale (PANSS) assessment, and magnetic resonance imaging scanning. Multivariate analyses of partial least squares (PLS) regression were used to explore the associations between whole-brain fractional amplitude of low-frequency fluctuations (fALFF), happy-neutral TSR (target pseudo-sentences were uttered in happy and neutral prosodies), and five PANSS factor scores (excitement/hostility, depression/anxiety, cognition, positive, and negative). ResultsThe happy prosody did not alter TSR or TSR changing rates in either SCHs or HCs. SCHs exhibited lower happy and neutral TSR than HCs. A fALFF PLS component (including precentral/postcentral gyrus, Subcallosal Cortex, several temporal regions, and cerebellum) was associated with happy and neutral TSR. SCHs demonstrated higher PLS fALFF scores and PLS TSR scores than HCs. In SCHs, PLS fALFF scores were correlated with the PANSS positive factor score, and PLS TSR scores were correlated with the PANSS cognition factor score. ConclusionsThe positive-psychiatric-symptoms-related spontaneous activity profile was associated with happy and neutral TSR, contributing to the cognition psychiatric symptoms dimension. The findings suggest the potential to improve positive and cognitive symptoms by enhancing happy and neutral TSR in schizophrenia based on neuroplasticity.