Speech Analysis Research Articles

Sentiment Analysis of Hate Speech on Women in Social Media Platform Using Multi Label Classification

Sentiment Analysis of Hate Speech on Women in Social Media Platform Using Multi Label Classification

Research Article Abstracts in Management Sciences Written by Thai EFL Scholars: An Analysis of Rhetorical Moves, Verb Tenses and Voices

Research article (RA) abstracts play a critical role in disseminating research findings, particularly challenging for non-native English speakers. While prior studies have explored the move analysis of RA abstracts across various subjects, the management sciences—a globally prominent field—have received less attention. This study addresses this gap by examining the rhetorical move patterns, verb tense usage, and voice constructions in 200 RA abstracts from Thai tier 1 management journals written by Thai scholars. The results showed that: 1) the Purpose-Method-Product structure was the most commonly employed move pattern; 2) present simple tense was mostly found in the Introduction, Purpose, and Conclusion sections, while past simple tense was preferred in the Method and Result sections; and 3) active voice was frequently used in the Introduction, Purpose, Product, and Conclusion sections, with a mix of active and passive voice mainly appearing in the Method section. The findings of the present study offer valuable insights for academic writing course developers, material designers, lecturers, and novice researchers, enhancing their understanding of RA abstract writing and supporting their integration into scientific communities.

High frequency post-pause word choices and task-dependent speech behavior characterize connected speech in individuals with mild cognitive impairment.

Alzheimer's disease (AD) is characterized by progressive cognitive decline, including impairments in speech production and fluency. Mild cognitive impairment (MCI), a prodrome of AD, has also been linked with changes in speech behavior but to a more subtle degree. This study aimed to investigate whether speech behavior immediately following both filled and unfilled pauses (post-pause speech behavior) differs between individuals with MCI and healthy controls (HCs), and how these differences are influenced by the cognitive demands of various speech tasks. Transcribed speech samples were analyzed from both groups across different tasks, including immediate and delayed narrative recall, picture descriptions, and free responses. Key metrics including lexical and syntactic complexity, lexical frequency and diversity, and part of speech usage, both overall and post-pause, were examined. Significant differences in pause usage were observed between groups, with a higher incidence and longer latencies following these pauses in the MCI group. Lexical frequency following filled pauses was higher among MCI participants in the free response task but not in other tasks, potentially due to the relative cognitive load of the tasks. The immediate recall task was most useful at differentiating between groups. Predictive analyses utilizing random forest classifiers demonstrated high specificity in using speech behavior metrics to differentiate between MCI and HCs. Speech behavior following pauses differs between MCI participants and healthy controls, with these differences being influenced by the cognitive demands of the speech tasks. These post-pause speech metrics can be easily integrated into existing speech analysis paradigms.

A Comparative Analysis of American and Vietnamese Presidents' Speeches: A Systemic Functional Grammar Perspective

This study presents a comparative analysis of the inaugural speeches of U.S. President Joe Biden and Vietnamese President To Lam, utilizing Michael Halliday's Systemic Functional Grammar (SFG) framework. By examining the Ideational (Transitivity), Interpersonal (Modality), and Textual (Texture and Cohesion) metafunctions, the study reveals how each leader's linguistic choices mirror the distinct political and cultural contexts of their respective nations. Joe Biden’s speech underscores the urgency of immediate action and crisis management, employing Material Processes to stress the importance of unity and national recovery. Conversely, To Lam's speech is rooted in long-term nation-building and the leadership of the Communist Party, with a focus on Relational Processes to reinforce collective responsibility and national unity. The Modality analysis demonstrates Joe Biden’s use of High Modality to convey decisiveness and transparency, aligning with American expectations for solid and clear commitments. In contrast, To Lam emphasizes collective duty and the Party's enduring leadership, reflecting Vietnam's socio-political ethos. Additionally, both speeches use Theme-Rheme structures and lexical repetition to ensure textual cohesion, though their cultural and ideological differences lead to distinct rhetorical approaches. By deepening our understanding of how political leaders use language to shape identities, communicate visions, and influence public perception, this study contributes to the broader field of political communication, offering insights into the strategic use of language across different cultural contexts.

The voice of depression: speech features as biomarkers for major depressive disorder.

Psychiatry faces a challenge due to the lack of objective biomarkers, as current assessments are based on subjective evaluations. Automated speech analysis shows promise in detecting symptom severity in depressed patients. This project aimed to identify discriminating speech features between patients with major depressive disorder (MDD) and healthy controls (HCs) by examining associations with symptom severity measures. Forty-four MDD patients from the Psychiatry Department, University Hospital Aachen, Germany and fifty-two HCs were recruited. Participants described positive and negative life events, which were recorded for analysis. The Beck Depression Inventory (BDI-II) and the Hamilton Rating Scale for Depression gauged depression severity. Transcribed audio recordings underwent feature extraction, including acoustics, speech rate, and content. Machine learning models including speech features and neuropsychological assessments, were used to differentiate between the MDD patients and HCs. Acoustic variables such as pitch and loudness differed significantly between the MDD patients and HCs (effect sizes 𝜼2 between 0.183 and 0.3, p < 0.001). Furthermore, variables pertaining to temporality, lexical richness, and speech sentiment displayed moderate to high effect sizes (𝜼2 between 0.062 and 0.143, p < 0.02). A support vector machine (SVM) model based on 10 acoustic features showed a high performance (AUC = 0.93) in differentiating between HCs and patients with MDD, comparable to an SVM based on the BDI-II (AUC = 0.99, p = 0.01). This study identified robust speech features associated with MDD. A machine learning model based on speech features yielded similar results to an established pen-and-paper depression assessment. In the future, these findings may shape voice-based biomarkers, enhancing clinical diagnosis and MDD monitoring.

Personalized Emotion Detection Adapting Models to Individual Emotional Expressions

Emotion recognition from text and speech has become a critical area of research in artificial intelligence (AI), enhancing human-computer interaction across various sectors. This paper explores the methodologies used in emotion recognition, focusing on Natural Language Processing (NLP) for text and acoustic analysis for speech. It reviews key machine learning and deep learning models, including Support Vector Machines (SVM), neural networks, and transformers, and highlights the datasets commonly used in emotion detection studies. The paper also addresses challenges such as multimodal integration, data ambiguity, and ethical considerations like privacy concerns and bias in models. Applications in customer service, healthcare, education, and entertainment are discussed, showcasing the growing importance of emotion recognition in AI- driven systems. Future research directions, including advancements in deep learning, multimodal systems, and real-time processing, are also explored to address existing limitations.

Comprehensive cost analysis of voice call recording storage with and without NAS and GSM gateway integration via unified communication server

This study is an extension of existing research, which previously demonstrated the effective integration of Network Attached Storage (NAS) with a GSM Gateway into Unified Communication (UC) Platform, thereby realizing remarkable enhancements in system scalability, data security, privacy, network and overall server performance. Although the technical and operational benefits of this integration were well understood and documented, however there was an important gap that is cost analysis. This paper addresses the gap by providing a comprehensive cost analysis of voice call archiving with and without NAS and GSM Gateway integration using UC. The analysis considers essential financial factors, including initial investment, ongoing operational costs, and long-term savings due to lower maintenance and scalability. While the findings reveal that the upfront costs of NAS and GSM Gateway integration are higher, these are offset by long-term operational efficiencies and system resilience, leading to substantial financial gains. The evaluation results offer organizations a strategic model to assess the cost-effectiveness of using this NAS-GSM Gateway integrated approach for voice data archiving in UC environments. This custom-costing solution provides a much-needed holistic perspective on the combined technical and financial benefits.

Clashing Narratives: An Ideological Square Analysis of Daniel Hagari and Abu Obaida Speeches

This study investigates the use of positive self-representation and negative other-representation in the speeches of two spokespersons, Daniel Hagari of Israel and Abu Obaida of Hamas, during the conflict that escalated after Operation Al-Aqsa Flood on October 7th, 2023. Utilizing Van Dijk's framework for discursive strategies, the analysis focuses on micro-level strategies such as actor description, comparison, consensus, evidentiality, national self-glorification, presupposition, and victimization. Findings reveal that while both spokespersons employed similar strategies to shape public perception, Daniel Hagari heavily utilized the comparison strategy to highlight the moral superiority of the IDF over Hamas. In contrast, Abu Obaida frequently used national self-glorification to bolster the morale of his audience and emphasize the righteousness of Hamas's cause. The study underscores the role of strategic discourse in conflict situations and calls for greater awareness and research into the linguistic tools used to propagate ideologies during war.

Polish Speech and Text Emotion Recognition in a Multimodal Emotion Analysis System

Emotion recognition by social robots is a serious challenge because sometimes people also do not cope with it. It is important to use information about emotions from all possible sources: facial expression, speech, or reactions occurring in the body. Therefore, a multimodal emotion recognition system was introduced, which includes the indicated sources of information and deep learning algorithms for emotion recognition. An important part of this system includes the speech analysis module, which was decided to be divided into two tracks: speech and text. An additional condition is the target language of communication, Polish, for which the number of datasets and methods is very limited. The work shows that emotion recognition using a single source—text or speech—can lead to low accuracy of the recognized emotion. It was therefore decided to compare English and Polish datasets and the latest deep learning methods in speech emotion recognition using Mel spectrograms. The most accurate LSTM models were evaluated on the English set and the Polish nEMO set, demonstrating high efficiency of emotion recognition in the case of Polish data. The conducted research is a key element in the development of a decision-making algorithm for several emotion recognition modules in a multimodal system.

Feasibility of Big Data Analytics to Assess Personality Based on Voice Analysis

Feasibility of Big Data Analytics to Assess Personality Based on Voice Analysis

Exploiting speech tremors: machine learning for early diagnosis of amyotrophic lateral sclerosis

Abstract Neurodegenerative diseases pose significant challenges in healthcare, with Amyotrophic Lateral Sclerosis (ALS) being one such rare yet debilitating condition affecting motor neurons. Machine learning (ML) and artificial intelligence (AI) have emerged as powerful tools in healthcare, offering insights and solutions for various medical conditions. This study investigates the application of ML to enhance early ALS diagnosis through the analysis of tremors in sustained speech. By focusing on tremor detection as a diagnostic marker, the research employs ML algorithms to develop predictive models capable of distinguishing ALS patients from healthy controls. The dataset comprises 54 patients from the Republican Research and Clinical Centre of Neurology and Neurosurgery in Belarus, Minsk. The study adopts a two-faceted approach: (1) Exploratory voice analysis to identify tremors associated with ALS in speech samples. (2) Development of ML algorithms to construct predictive models for early ALS diagnosis based on the identified tremors. The ML models exhibit promising results in distinguishing ALS patients from healthy controls based on speech analysis. Tremor detection in sustained speech proves to be an effective marker for early ALS diagnosis. While initial findings are encouraging, larger-scale studies are required to validate the clinical applicability of this approach. The successful application of ML and AI in early ALS diagnosis by leveraging innovative approaches, such as tremor detection in sustained speech, we can enhance early diagnosis and improve patient outcomes in neurodegenerative diseases like ALS on a broader scale.

Reply: Enhancing automated speech analysis for hepatic encephalopathy detection.

Reply: Enhancing automated speech analysis for hepatic encephalopathy detection.

Letter to the Editor: Enhancing automated speech analysis for hepatic encephalopathy detection.

Letter to the Editor: Enhancing automated speech analysis for hepatic encephalopathy detection.

PHONETIC MEANS OF EXPRESSING NEGATION IN MODERN BULGARIAN

The object of the current study are the phonetic means that express negation in Contemporary Bulgarian and that are part of the formal inventory of the functional and semantic category of polarity (and more precisely – part of the periphery of the negative polarity). Our interest on this topic was promoted by the lack of detailed research. The main goal here is to present a description of these specific characteristics by paying more attention to the role of intonation and stress in the speech realization of the phrases of interest. To achieve this goal, we have employed a variety of technological resources for rhythmic and melodic analysis, with the help of which we have conducted experimental research on our targeted phrases and provided graphic representation of melody and intensity of pronunciation, as well as spectrographic analysis of speech. Our data shows that stress and intonation are indeed differentiating factors in the meaning of the phonetic explicators we observe. Stress and intonation make t

A pragmatic analysis of manipulative speech acts used by American and Arabic attorneys in selected criminal trials

The present study investigates the manipulative speech acts used by American and Arabic attorneys while defending their clients in criminal trials. It is hypothesized that they construct their arguments on the basis of such type of speech acts to mislead the members of the jury who expect to listen to concrete evidences of innocence. The study follows a qualitative analysis to achieve the aims in terms of the modal adopted. The model is applied to six selected arguments divided between American and Arabic attorneys. Each part is resembled by an attorney. The results show that they both use indirect speech acts of manipulation, with some notable differences. The conclusions are drawn on the basis of their manipulative speech acts and on the basis of the comparison between the two attorneys which prove that they resort to two different types of indirect speech acts to manipulate the members of the jury.

Analysis of vocal characteristics in hypopharyngeal cancer patients with vocal cord dysfunction

Objective:To analyze the acoustic characteristics of patients with hypopharyngeal cancer accompanied by vocal cord dysfunction. Methods:A retrospective analysis was conducted on the data of patients with hypopharyngeal cancer who were initially treated at The Second Affiliated Hospital of Xi'an Jiaotong University from January 2018 to April 2024. Patients who had completed electronic laryngoscopy, stroboscopic laryngoscopy, and voice analysis were selected from the data. Among them, patients with hypopharyngeal cancer who had unilateral vocal cord activity disorders were selected as the experimental group, while patients with symmetrical bilateral vocal cord activity were assigned to the control group. Then the clinical characteristics, the vocal parameters, and the stroboscopic laryngoscopy results of patients with hypopharyngeal cancer in the experimental group and the control group were analyzed and compared. Results:Compared with that in the control group, the proportion of lesions located on the inner wall of the piriform fossa in the experimental group increased（83.3% vs 53.8%）, and the difference was statistically significant（P<0.05）. There was no significant difference in vocal parameters such as SPL, Jitter, Shimmer, MPT, DSI, F0, sound intensity, electroglottic value and VHI between the experimental group and the control group（P>0.05）. However, the values of F0, Jitter, Shimmer and VHI in the experimental group were higher than those in the control group. In addition, in terms of the results of stroboscopic laryngoscopy, the proportion of glottic insufficiency（42.9% vs 18.8%） and asymmetric arytenoid cartilage（64.3% vs 0） in the experimental group was significantly higher than that in the control group（P<0.05）. However, the mucosal waves of the vocal cords on the affected side did not weaken in patients in both the experimental group and the control group. In the experimental group of 18 patients with hypopharyngeal cancer who received induction chemotherapy（nituzumab+nedaplatin+5-fluorouracil）, 13 of them had improved vocal cord activity（improvement rate of 72.2%）. Conclusion:Hypopharyngeal cancer in the medial wall of the pyriform fossa is more prone to vocal cord dysfunction, but vocal cord dysfunction has little effect on the vocal parameters of patients with hypopharyngeal cancer.

Beyond breathalyzers: AI-powered speech analysis for alcohol intoxication detection

Detecting potential alcohol inebriation or intoxication status holds paramount significance for social prevention and security. Beyond its association with long-term health effects, alcohol consumption can lead to immediate consequences, including reduced control over one’s actions, with traffic fatalities representing one of the most tragic outcomes.This study leveraged the Alcohol Language corpus, involving 162 subjects recorded both in sober and inebriated states. Participants provided 60 speech samples while sober and 30 when intoxicated, all within a realistic car setting using head-mounted microphones. Our research endeavors encompassed comprehensive stratified statistical tests to examine the impact of alcohol consumption on speech production while uncovering the influence of covariates such as age, gender, and drinking habits.Additionally, we introduced a speaker-neutral machine learning algorithm, based on the Domain-Adversarial Neural Network architecture. This approach aimed to overcome challenges posed by individual differences that often complicate intoxicated speech analysis. Notably, our findings highlighted the effectiveness of features like the RASTA-filtered auditory spectrum. Nevertheless, the results from statistical tests emphasized the need for techniques that minimize inter-subject variability.As for the automatic classification, the proposed architecture exhibited promising results, yielding a classification accuracy slightly exceeding 70% on an independent test set. Although preliminary, our research demonstrates the potential for detecting alcohol-induced speech changes, benefiting societal well-being and security. It also underscores the importance of developing strategies that account for individual differences while harnessing the power of automatic models to effectively distinguish between sober and intoxicated individuals.

A process-oriented analysis of speech and silent intervals in responses to serial naming tasks.

A process-oriented analysis of speech and silent intervals in responses to serial naming tasks.

Development of cosmoTA as an Analysis Tool for Children Speech Analysis

Development of cosmoTA as an Analysis Tool for Children Speech Analysis

Analisis semiotika pidato kenegaraan terakhir Presiden Joko Widodo pada sidang tahunan MPR 2024

This research aims to examine the semiotic analysis of President Jokowi's final state speech at the 2024 MPR annual session through Charles Sanders Pierce's semiotic approach which involves the analysis of representamen (signs), objects (sign references), and interpretants (use of signs). This research using descriptive qualitative methods to understand the meaning and content contained in the speech. The results of the study show that "Semiotic analysis of President Jokowi's final state speech at the 2024 MPR annual session" effectively conveys emotional messages about apologies, conveying achievements, and leadership hopes. in the future. This speech teaches that whatever we do there must be something good and not good. As humans we need to be humble and apologize for what we have done so far