-
PloS One 2024Digital speech recognition is a challenging problem that requires the ability to learn complex signal characteristics such as frequency, pitch, intensity, timbre, and...
Digital speech recognition is a challenging problem that requires the ability to learn complex signal characteristics such as frequency, pitch, intensity, timbre, and melody, which traditional methods often face issues in recognizing. This article introduces three solutions based on convolutional neural networks (CNN) to solve the problem: 1D-CNN is designed to learn directly from digital data; 2DS-CNN and 2DM-CNN have a more complex architecture, transferring raw waveform into transformed images using Fourier transform to learn essential features. Experimental results on four large data sets, containing 30,000 samples for each, show that the three proposed models achieve superior performance compared to well-known models such as GoogLeNet and AlexNet, with the best accuracy of 95.87%, 99.65%, and 99.76%, respectively. With 5-10% higher performance than other models, the proposed solution has demonstrated the ability to effectively learn features, improve recognition accuracy and speed, and open up the potential for broad applications in virtual assistants, medical recording, and voice commands.
Topics: Neural Networks, Computer; Humans; Speech Recognition Software; Speech; Algorithms
PubMed: 38669233
DOI: 10.1371/journal.pone.0302394 -
Behavioral Sciences (Basel, Switzerland) Apr 2024Danish prosecutors report exposure to negative acts from professional counterparts in courtrooms, which is associated with an increased risk of burnout. However,...
Danish prosecutors report exposure to negative acts from professional counterparts in courtrooms, which is associated with an increased risk of burnout. However, knowledge of the characteristics of these acts is limited. Based on existing theoretical frameworks, this study aims to characterize these negative acts. A nation-wide survey of Danish prosecutors (response rate: 81%) yielded 687 descriptions of experiences with negative acts from professional counterparts from a career perspective. These were analyzed using theory-directed content analysis based on the Stress-as-Offense-to-Self (SOS) theory by Semmer and colleagues and Cortina and colleagues' characterization of incivility in American courtrooms. We identified a total of 15 types of behavior within the three main themes: illegitimate tasks (n = 22), illegitimate stressors (n = 68), and illegitimate behavior (n = 612). Tentative differences in the distribution of experienced negative acts from a career perspective were found for gender and seniority. Women reported negative acts more frequently than men, and assistant prosecutors reported verbal abuse more frequently than senior prosecutors, who, conversely, more often reported a perceived lack of court management. More prospective research is needed on negative acts experienced by prosecutors to assess the scope of these in Danish courtrooms and how they impact the risk of burnout.
PubMed: 38667128
DOI: 10.3390/bs14040332 -
Alzheimer's Research & Therapy Apr 2024Plasma neurofilament light chain (NfL) is a promising biomarker of neurodegeneration with potential clinical utility in monitoring the progression of neurodegenerative...
BACKGROUND
Plasma neurofilament light chain (NfL) is a promising biomarker of neurodegeneration with potential clinical utility in monitoring the progression of neurodegenerative diseases. However, the cross-sectional associations of plasma NfL with measures of cognition and brain have been inconsistent in community-dwelling populations.
METHODS
We examined these associations in a large community-dwelling sample of early old age men (N = 969, mean age = 67.57 years, range = 61-73 years), who are either cognitively unimpaired (CU) or with mild cognitive impairment (MCI). Specifically, we investigated five cognitive domains (executive function, episodic memory, verbal fluency, processing speed, visual-spatial ability), as well as neuroimaging measures of gray and white matter.
RESULTS
After adjusting for age, health status, and young adult general cognitive ability, plasma NfL level was only significantly associated with processing speed and white matter hyperintensity (WMH) volume, but not with other cognitive or neuroimaging measures. The association with processing speed was driven by individuals with MCI, as it was not detected in CU individuals.
CONCLUSIONS
These results suggest that in early old age men without dementia, plasma NfL does not appear to be sensitive to cross-sectional individual differences in most domains of cognition or neuroimaging measures of gray and white matter. The revealed plasma NfL associations were limited to WMH for all participants and processing speed only within the MCI cohort. Importantly, considering cognitive status in community-based samples will better inform the interpretation of the relationships of plasma NfL with cognition and brain and may help resolve mixed findings in the literature.
Topics: Humans; Male; Neurofilament Proteins; Aged; Middle Aged; Independent Living; Cross-Sectional Studies; Cognitive Dysfunction; Neuroimaging; Cognition; Biomarkers; Neuropsychological Tests; Magnetic Resonance Imaging; Brain; White Matter; Aging
PubMed: 38664843
DOI: 10.1186/s13195-024-01464-1 -
Scientific Reports Apr 2024Depression, a pervasive global mental disorder, profoundly impacts daily lives. Despite numerous deep learning studies focused on depression detection through speech...
Depression, a pervasive global mental disorder, profoundly impacts daily lives. Despite numerous deep learning studies focused on depression detection through speech analysis, the shortage of annotated bulk samples hampers the development of effective models. In response to this challenge, our research introduces a transfer learning approach for detecting depression in speech, aiming to overcome constraints imposed by limited resources. In the context of feature representation, we obtain depression-related features by fine-tuning wav2vec 2.0. By integrating 1D-CNN and attention pooling structures, we generate advanced features at the segment level, thereby enhancing the model's capability to capture temporal relationships within audio frames. In the realm of prediction results, we integrate LSTM and self-attention mechanisms. This incorporation assigns greater weights to segments associated with depression, thereby augmenting the model's discernment of depression-related information. The experimental results indicate that our model has achieved impressive F1 scores, reaching 79% on the DAIC-WOZ dataset and 90.53% on the CMDC dataset. It outperforms recent baseline models in the field of speech-based depression detection. This provides a promising solution for effective depression detection in low-resource environments.
Topics: Humans; Depression; Speech; Deep Learning; Neural Networks, Computer
PubMed: 38664511
DOI: 10.1038/s41598-024-60278-1 -
International Journal of Law and... 2024Verbal and physical violence in psychiatric hospitals can have harmful consequences for staff members, such as physical injury, traumatisation, and sick leave, and they...
BACKGROUND
Verbal and physical violence in psychiatric hospitals can have harmful consequences for staff members, such as physical injury, traumatisation, and sick leave, and they often accompany involuntary admission. Harm to others may co-occur with self-harm, i.e., dual harm. However, little is known about the association between dual-harm and violent behaviour towards staff members and its clinical outcomes, such as seclusion and rapid tranquilisation after involuntary admission to a psychiatric inpatient unit.
METHOD
A convenience sample of patients admitted involuntarily (N = 384; mean age = 48.03, SD = 19.92) between January 2016 and December 2019 in Western Brabant, the Netherlands, was used to design a retrospective file audit. Distinct harm groups, marked by the presence/absence of self- and/or other-harm, were investigated using multivariate linear regression modelling on the seriousness of violent acts and the total length of admission. Logistic regression analyses were used to study the association between harm groups and the administration of rapid tranquilisation, seclusion, and extended involuntary admissions.
RESULTS
Several harm groups were identified, including self-harm only, other-harm only, and dual-harm groups. Psychiatric patients admitted to the hospital because of (the risk of) violence towards others had a higher risk of violent incidents during admission and some restrictive measures. In a subgroup of patients with psychotic disorders, patients with dual harm committed the most serious violent incidents compared to those in the other harm groups.
CONCLUSION
Distinct harm groups were identified in a sample of involuntarily admitted patients. In a general adult psychiatric setting, patients at risk for violent behaviour, especially dual-harm patients, should be identified and monitored as part of the risk assessment. Future research is needed to explore more clinical correlates in the proposed distinction between harmful groups and to assess long-term prognosis.
Topics: Humans; Netherlands; Male; Female; Self-Injurious Behavior; Middle Aged; Adult; Hospitals, Psychiatric; Retrospective Studies; Violence; Mental Disorders; Commitment of Mentally Ill; Aged
PubMed: 38663172
DOI: 10.1016/j.ijlp.2024.101989 -
Frontiers in Psychiatry 2024The purpose of the study was to determine how Adverse Childhood Experiences (ACE) relate to adulthood flourishing, symptoms of depression, anxiety, somatization,...
The purpose of the study was to determine how Adverse Childhood Experiences (ACE) relate to adulthood flourishing, symptoms of depression, anxiety, somatization, self-reported health, sexual risk behaviors, and alcohol consumption. A quantitative cross-sectional methodology was used. A total of 452 adults completed the survey. The most prevalent ACE include physical abuse (44.69%), separation/divorce of parents (41.81%), living with someone with alcohol problems (39.38%), and being sworn, insulted, or humiliated by adults at home (35.62%). Almost one out of every four respondents (24.34%) reported being touched by an adult, 17.92% reported that an adult tried to manipulate the respondent into touching them, and 8.19% were forced to have sexual intercourse. Results indicate that women reported a higher number of ACE than men. The number of ACE is inversely related to flourishing and self-reported health; while being positively associated with participant's scores in depression, anxiety, somatization, sexual risk behaviors, and alcohol use. The regression model, including the eleven ACE and respondents' sex and age, achieved medium effect sizes for somatization, depression, and anxiety symptoms and small effect sizes for flourishing, self-reported health, sexual risk behaviors, and alcohol consumption. Specific ACE have a particularly significant negative impact on mental health outcomes: forced intercourse, witnessing familial violence, verbal humiliation, and living with individuals struggling with mental health issues and drug consumption or who were incarcerated. In conclusion, the study highlights the alarming prevalence of ACE among the Honduran population and their significant negative impact on mental health outcomes during adulthood.
PubMed: 38659462
DOI: 10.3389/fpsyt.2024.1352824 -
Scientific Reports Apr 2024This work presents data from 148 German native speakers (20-55 years of age), who completed several speaking tasks, ranging from formal tests such as word production...
This work presents data from 148 German native speakers (20-55 years of age), who completed several speaking tasks, ranging from formal tests such as word production tests to more ecologically valid spontaneous tasks that were designed to mimic natural speech. This speech data is supplemented by performance measures on several standardised, computer-based executive functioning (EF) tests covering domains of working-memory, cognitive flexibility, inhibition, and attention. The speech and EF data are further complemented by a rich collection of demographic data that documents education level, family status, and physical and psychological well-being. Additionally, the dataset includes information of the participants' hormone levels (cortisol, progesterone, oestradiol, and testosterone) at the time of testing. This dataset is thus a carefully curated, expansive collection of data that spans over different EF domains and includes both formal speaking tests as well as spontaneous speaking tasks, supplemented by valuable phenotypical information. This will thus provide the unique opportunity to perform a variety of analyses in the context of speech, EF, and inter-individual differences, and to our knowledge is the first of its kind in the German language. We refer to this dataset as SpEx since it combines speech and executive functioning data. Researchers interested in conducting exploratory or hypothesis-driven analyses in the field of individual differences in language and executive functioning, are encouraged to request access to this resource. Applicants will then be provided with an encrypted version of the data which can be downloaded.
Topics: Humans; Executive Function; Adult; Middle Aged; Female; Male; Speech; Germany; Young Adult; Language; Memory, Short-Term; Neuropsychological Tests
PubMed: 38658576
DOI: 10.1038/s41598-024-58617-3 -
ENeuro May 2024More and more patients worldwide are diagnosed with dementia, which emphasizes the urgent need for early detection markers. In this study, we built on the auditory...
More and more patients worldwide are diagnosed with dementia, which emphasizes the urgent need for early detection markers. In this study, we built on the auditory hypersensitivity theory of a previous study-which postulated that responses to auditory input in the subcortex as well as cortex are enhanced in cognitive decline-and examined auditory encoding of natural continuous speech at both neural levels for its indicative potential for cognitive decline. We recruited study participants aged 60 years and older, who were divided into two groups based on the Montreal Cognitive Assessment, one group with low scores ( = 19, participants with signs of cognitive decline) and a control group ( = 25). Participants completed an audiometric assessment and then we recorded their electroencephalography while they listened to an audiobook and click sounds. We derived temporal response functions and evoked potentials from the data and examined response amplitudes for their potential to predict cognitive decline, controlling for hearing ability and age. Contrary to our expectations, no evidence of auditory hypersensitivity was observed in participants with signs of cognitive decline; response amplitudes were comparable in both cognitive groups. Moreover, the combination of response amplitudes showed no predictive value for cognitive decline. These results challenge the proposed hypothesis and emphasize the need for further research to identify reliable auditory markers for the early detection of cognitive decline.
Topics: Humans; Female; Male; Aged; Cognitive Dysfunction; Electroencephalography; Middle Aged; Evoked Potentials, Auditory; Speech Perception; Aged, 80 and over; Cerebral Cortex; Acoustic Stimulation; Speech
PubMed: 38658138
DOI: 10.1523/ENEURO.0545-23.2024 -
Trends in Hearing 2024Negativity bias is a cognitive bias that results in negative events being perceptually more salient than positive ones. For hearing care, this means that hearing aid... (Randomized Controlled Trial)
Randomized Controlled Trial
Negativity bias is a cognitive bias that results in negative events being perceptually more salient than positive ones. For hearing care, this means that hearing aid benefits can potentially be overshadowed by adverse experiences. Research has shown that sustaining focus on positive experiences has the potential to mitigate negativity bias. The purpose of the current study was to investigate whether a positive focus (PF) intervention can improve speech-in-noise abilities for experienced hearing aid users. Thirty participants were randomly allocated to a control or PF group (N = 2 × 15). Prior to hearing aid fitting, all participants filled out the short form of the Speech, Spatial and Qualities of Hearing scale (SSQ12) based on their own hearing aids. At the first visit, they were fitted with study hearing aids, and speech-in-noise testing was performed. Both groups then wore the study hearing aids for two weeks and sent daily text messages reporting hours of hearing aid use to an experimenter. In addition, the PF group was instructed to focus on positive listening experiences and to also report them in the daily text messages. After the 2-week trial, all participants filled out the SSQ12 questionnaire based on the study hearing aids and completed the speech-in-noise testing again. Speech-in-noise performance and SSQ12 Qualities score were improved for the PF group but not for the control group. This finding indicates that the PF intervention can improve subjective and objective hearing aid benefits.
Topics: Humans; Hearing Aids; Male; Female; Speech Intelligibility; Speech Perception; Aged; Noise; Middle Aged; Correction of Hearing Impairment; Persons With Hearing Impairments; Perceptual Masking; Hearing Loss; Audiometry, Speech; Surveys and Questionnaires; Aged, 80 and over; Time Factors; Acoustic Stimulation; Hearing; Treatment Outcome
PubMed: 38656770
DOI: 10.1177/23312165241246616 -
Acta Psychologica Jun 2024Several studies have shown that anxious individuals experience a slower passage of time under threat conditioning. Anxiety-evoking situations have also been proposed to...
Several studies have shown that anxious individuals experience a slower passage of time under threat conditioning. Anxiety-evoking situations have also been proposed to elevate arousal levels, which, in turn, alter one's time percept. However, the effect of social stressors on time perception remains significantly neglected. The current research aimed to investigate the impact of anxiety levels on time estimation and passage of time judgments during public speaking in healthy adults. Participants were recruited from a pool of students that had to give a presentation as part of a university course or their teaching duties. Following the presentation, they were asked to make retrospective time estimations on the duration of the latter, as well as to provide passage of time judgments. Self-reported questionnaires related to affective states, public speaking anxiety, and performance were also administered. Analysis showed that higher levels of public speaking anxiety predicted temporal overestimation and slower "feel" duration and passage of time. Moreover, the relationship between public speaking anxiety and passage of time was mediated by participants' mood states, which remained significant after -indirectly- controlling for fear of evaluation. Overall, our observations suggest that anxiety levels during public presentation significantly predict altered perception and experience of time. The latter can be explained by the speaker's mood status. Identifying the mechanisms that modulate timing under psychological stressors could complement our understanding regarding their impact on educational and social settings, as well as set the ground towards the development of early intervention and prevention strategies for those who suffer from stress-related disorders.
Topics: Humans; Male; Female; Adult; Speech; Time Perception; Anxiety; Young Adult; Affect; Stress, Psychological; Adolescent
PubMed: 38653079
DOI: 10.1016/j.actpsy.2024.104268