Audio Content Research Articles

Can media inspire public engagement with astronomy? Assessing information modalities and potential mechanisms for inspiration in a basic science context

Although many science communicators strive to inspire audiences, scant research has sought to understand how media may evoke inspiration. The present study was a three-condition (modality: text-only, audio-only, and audiovisual) between-subjects experiment examining how media content about the James Webb Space Telescope (JWST) motivated participants’ feeling moved, awe, curiosity, rumination, and their inspiration state. The findings revealed no statistical difference between using text or audio content, but that audiovisual content can hinder rumination if self-transcendent emotions are not induced. Findings revealed that content leading participants to feeling moved and experiencing awe should lead to reflective thought, and ultimately, feelings of inspiration.

I-vector and variability compensation techniques for mobile phone recognition

Mobile phone recognition consists of trying to identify the mobile phone brand or model, which is very important in forensic analysis. In this paper, we exploit the audio recordings to realize this task. Audio data conveys many information, one of them is related to the source of records, i.e., is a specific audio was captured by a given mobile phone. In order to separate between this particular information and the rest of audio content, we introduce the I-vector technique. Additionally, Linear Discriminant Analysis (LDA) and Within-Class Covariance Normalization (WCCN) methods were used in different ways to compensate the speaker and channel variability. We evaluate the performance of our proposed system by correct rate recognition based on Gaussian Probabilistic Linear Discriminant Analysis (GPLDA) and Cosine Similarity Score (CSS) methods. To train the mobile phone models and do tests, two datasets were prepared: the public MOBIPHONE database and a novel local database. Experiments done with different configurations demonstrate the effectiveness of I-vectors models with the combination of LDA – WCCN compensation techniques and CSS as a similarity measurement method. A correct recognition rates of 97.23 %, 98.91 %, and 99.45 % were obtained for MOBIPHONE, the local Same Environment (SE) and Different Environment (DE) databases, respectively.

Students’ Perception on The Use of TED Talks to Improve Listening Skills

Enhancing listening skills is crucial for junior high school students learning English. Innovative teaching tools such as TED Talks offer an interactive platform that might aid this educational goal. This research aims to evaluate the effect of TED Talks on developing listening skills among junior high school learners. The study employed a mixed-methods design involving 34 junior high school students. Participants underwent pre-and post-test assessments to measure their listening skills. Additionally, interviews were conducted to delve deeper into the students' experiences and perspectives regarding using TED Talks in their learning process. The findings revealed that TED Talks significantly enhanced students' English listening skills. After engaging with TED Talks videos, students showed notable improvements in vocabulary acquisition, pronunciation accuracy, and listening comprehension. The high-quality audio and diverse content of TED Talks were credited for these improvements. Furthermore, interviews provided insights into the positive impact of TED Talks on student motivation and engagement, with students expressing increased satisfaction and perceived efficacy in the classroom. TED Talks also contributed to the acquisition of advanced vocabulary and sharpened students' critical thinking skills through exposure to various topics.

Evolution and future directions of Artificial Intelligence Generated Content (AIGC): A comprehensive review

Artificial Intelligence Generated Content (AIGC) has rapidly evolved, revolutionizing the creation of text, images, audio, and video content. Despite these advancements, research on the development process of AIGC technology remains scarce, necessitating a systematic discussion of its current state and future directions. So this paper delves into the significant advancements and foundational technologies driving AIGC, emphasizing the contributions of state-of-the-art models such as DALL-E 3 [1] and Sora [2]. We analyze the evolution of generative models from single-modal approaches to the current multimodal generative models. The paper further explores the application prospects of AIGC across various domains such as office work, art, education, and film, while addressing the existing limitations and challenges in the field. We propose potential improvement directions, including more efficient model architectures and enhanced multimodal capabilities. Emphasis is placed on the environmental impact of AIGC technologies and the need for sustainable practices. Our comprehensive review aims to provide researchers and professionals with a deeper understanding of AIGC, inspiring further exploration and innovation in this transformative domain.

Customer Satisfaction towards OTT Platform

Over the top platform has grown drastically over a period of time. The platform delivers online video and audio contents. In this regard it becomes important to study the customer satisfaction towards OTT platforms. The study has been conducted by taking in to consideration six dimensions namely content quality, user interface, personalization and recommendation, price and value, customer service and support, accessibility and convenience. Each dimension is having few parameters. The study is purely primary in nature, the data has been collected through structured questionnaire from 200 respondents using convenience sampling technique and has been analysed using SPSS software. Anova test has been used for data analysis. The research is confined to only Ballari city. The results obtained from the analysis shows that all the dimensions are having positive association towards customer satisfaction.

Exploring Podcast And Videocast Consumption: A Construct Development And Validation Study Using TPB

Digital content consumption, particularly through podcasts and videocasts, is rapidly reshaping media engagement patterns globally. While existing research has explored general online behavior, there remains a gap in understanding the psychological drivers specific to digital audio and video content consumption, especially in emerging markets. This study aims developed and validated a Digital Content Consumption (DCC) scale within the Brazilian context, integrating it with the Theory of Planned Behavior (TPB) to enhance our understanding of consumption behavior. A cross-sectional survey was conducted among Brazilian consumers during thew month May of 2023. Participants were recruited through a convenience sample of adult internet users who had consumed podcasts or videocasts within the past month. The primary exposure was the consumption of digital audio and video content. The study employed Exploratory Factor Analysis, Confirmatory Factor Analysis, and Structural Equation Modeling (SEM) to analyze the data and validate the DCC scale. The final sample comprised 223 participants, with a mean age of 32 years and an even gender distribution. The SEM analysis demonstrated a strong model fit for the DCB scale within the TPB framework. Attitudes emerged as a significant predictor of both intention and behavior in digital content consumption. Utilitarianism was found to mediate the relationship between attitudes and consumption. The DCCB scale showed high internal consistency and construct validity across different demographic subgroups. The development and validation of the DCC scale provide a robust tool for measuring digital content consumption behavior. This study extends the application of the Theory of Planned Behavior to the specific domain of podcast and videocast consumption, offering valuable insights for content creators and marketers. Future research should explore the scale's applicability in diverse cultural contexts and its potential for predicting long-term engagement patterns with longitudinal studies

Cultural Aspects of the Transformation of t he Term Podcasting

This article presents the results of a study of the transformation of the very concept of podcasting in domestic scientifi c discourse, in the Russian media fi eld from a cultural point of view. Podcasting is a form of transmission of mainly audio or video materials (VODcasts) to the Internet, which has existed in the world for about twenty years. Today, podcasts are associated not only with spoken audio content, but also with video interviews, and even with voice messages on the social media Telegram. Moreover, a podcast can be understood as almost any work in the digital media space, even text. This work was carried out in 2023 as part of a comprehensive study of the transformation of the audiovisual environment on new technological platforms, which has been carried out by the Department of Television and Radio Broadcasting of the Faculty of Journalism of Moscow State University since 2017. The purpose of this stage was to determine what is currently understood by the word podcasting and an attempt to capture its distinctive format features, cultural aspects of the evolution of the concept. The methods used were conceptual and contextual types of analysis. The cognitive approach made it possible to consider the verbalized key concepts of scientifi c discourse. The defi nitions given to podcasting by domestic and foreign researchers, practitioners, market participants in the process of development and formation of the industry, as well as a control group of listeners were analyzed. The authors of this article make an attempt to give the most relevant cultural interpretation of the concept of podcasting. The main conclusion is that Russian-language podcasting, for a number of reasons, is developing along a trajectory that is largely different from the world one, in particular the American one. But its development is complicated by a number of factors, including geopolitical and cultural ones. The concept of “podcasting” will most likely continue its evolutionary movement in search of a more solid position in the sociocultural space. Keywords: podcasting, media system, audiovisual content, social media, broadcasting, formats, genres

Peculiarities of online user’s content search in the context of inbound marketing

The article addresses the issue of the distribution of users browsing the internet for content in a specific format through alternative content delivery channels in the context of inbound marketing. The problem being addressed is part of a broader problem with the formation of traffic to an organization’s website. The theoretical part of the work presents the concept of inbound marketing and analyzes the theoretical aspects of its tactical activities, such as content marketing, search engine marketing, and social media marketing. Based on the theoretical research, the scientific hypothesis is formulated – there are links between the desired content format and the choice of content search and consumption channel. The research methodology consists of data collection by survey method and hypothesis testing using Cochran’s Q test. Empirical research has shown links between the content format sought and the channels for searching and consuming content. Although in almost all cases the search for video and audio content is dominated by search engines, evaluating other content delivery channels reveals a significant specialty of users’ web browsing.

A Multidisciplinary Threats to Emerging Cybersecurity, Legal and Ethical Threats Posed by Deepfake Technology

The creation of remarkably realistic but phony audio and video content through the use of deepfake technology raises a number of issues with cybersecurity, law and ethics. Deep fakes undermine the credibility of online information sources by posing risks including identity theft and the dissemination of false information. Legally speaking, the rise of deepfakes highlights regulatory inadequacies and enforcement concerns while posing difficult considerations about privacy rights and responsibility. Due to the potential for reputational injury and the perpetuation of negative stereotypes, deep fakes raise ethical questions about permission, privacy and harm to society. Deep fakes have become more common as a result of artificial intelligence's quick development which calls for preventative steps to counter new dangers. To effectively secure against misuse and maintain the integrity of digital material, this research paper proposes for a multidisciplinary strategy that takes legal, ethical and cybersecurity concerns into account

Sarcasm Detection on Social Media: Addressing the Issues of Short and Long Texts Using Machine Learning and Deep Learning Approaches

Sarcasm is a linguistic term that conveys a meaning that is different from what is meant to be said, frequently used to mock, taunt, or convey disdain. Sarcasm is a complicated social phenomenon that can be recognized by its tone of voice, exaggeration, or context. Because sarcastic discourse is nuanced, detecting sarcasm in Natural Language Processing (NLP) has become a major difficulty. This paper thoroughly examines the tradition machine learning approaches like SVM, Naive Bayes and Random Forest well as advanced deep learning methods such as RNN, LSTM, and transformer based models, like BERT, that has demonstrated superior performance used for sarcasm detection, and also the paper offers more sophisticated data preprocessing techniques that comprise several stages, each focusing on a different facet of the fragmented and informal character of the social media material. It focuses mainly on short and long texts that may be found in news headlines and social media sites like Facebook, Instagram, X (formerly known as Twitter), and others. The paper also explores various challenges like cultural and linguistic barriers, and increased use of audio and visual sarcastic content on social media. At last paper concludes with possible guidelines for future works including the development of real-time, multilingual systems to examining holistic strategies that can encompass the complexity of sarcasm in multimodal communications. Key Words: sarcasm, NLP, social media, machine learning, deep learning

Modeling Android Security Vulnerabilities: Insights from Statistical Distributions

Android operating system is a mobile operating system that supports multimedia features. Android offers a wide range of applications and integrated features for playing, recording, editing and sharing audio, video, images and other multimedia content. Most Android devices include cameras, speakers, microphones, and other multimedia components. In software security, vulnerabilities are critical concerns that often emerge during software development. Predicting these vulnerabilities post-release is essential for risk assessment and mitigation. While various models have been explored, the Android operating system remains relatively uncharted. This study delves into modeling Android security vulnerabilities using different statistical distributions, comparing their suitability to the widely-used Alhazmi-Malaiya Logistic (AML) model. Data from the National Vulnerability Database (NVD) spanning 2016 to 2018, along with Common Vulnerability Scoring System (CVSS) scores, was analyzed. The study evaluates several distribution models, including Logistic, Weibull, Nakagami, Gamma, and Log-logistic, for monthly vulnerability counts and average monthly impact values. Goodness-of-fit tests and information criteria were applied for model robustness assessment. The findings offer valuable insights for researchers and Android software developers, aiding prediction, risk assessment, resource allocation, and research direction. Logistic and Nakagami distributions emerged as the best-fit models for average monthly impact values and monthly vulnerability counts, respectively. Finally, statistical methods perform better against known artificial intelligence methods for small data sets or more clearly defined data due to their flexible features such as comprehensibility, amount of data, need for calculation, and data independence.

Enhancing Museum Experiences: A Multi-Institution Mobile Multimedia Delivery System Using BLE Beacons

This paper presents a solution using Bluetooth Low Energy (BLE) beacon technology to enhance visitor experiences in museums, both indoors and outdoors. Organizations using the developed mobile service can associate text, images, video, and audio content to specific exhibits, providing a dynamic and engaging encounter for visitors. This innovative approach extends applicability to a wide range of organizations, not just museum environments.

Translanguaged practice in listening assessment: L1 vs. L2 responses in recall tasks

ABSTRACT Listening recall tasks, where test takers write L2 responses based on audio content, often inadvertently measure L2 writing skills, potentially compromising the accuracy of listening assessment. To address this concern, the present study draws inspiration from translanguaging, and employs a mixed-methods approach to investigate the potential of using L1 responses into recall tasks to mitigate the impact of L2 writing proficiency on task outcomes, thus enhancing task validity. Specifically, a listening recall task requiring L1 responses was developed and administered to 102 L2 learners in low and intermediate English proficiency. Their performance on this task was compared to a control task that required L2 responses. Follow-up interviews were also conducted. The results indicated that allowing L1 responses in the recall task led to better performance for intermediate-level L2 learners, enhanced representation of the listening construct, and garnered participants’ preference as a listening assessment approach. This suggests that utilising L1 responses has the potential to improve the construct and face validity of recall tasks, addressing the initial validity concern. Consequently, this investigation contributes to the refinement of listening task development and advocates for the application of translanguaged practices in language assessment, advancing the broader concept of translanguaging in language education.

TelMedAI

Telemedicine is the practice of technology-enabled remote communication between patient and doctor. This phenomenon in healthcare has the potential to make services affordable and save time and money. Besides telemedicine allows care givers and family members to join conversations with doctors. Indian government initiated the National Telemedicine Network (NTN) to serve remote areas in healthcare by integrating existing healthcare facilities.Literature has revealed that existing works lack in an integrated approach for patient speech translation in language-independent fashion and automatic detection of disease and symptoms based on speech.There is a need for an automated system using Artificial Intelligence (AI) to recognize patient's speech and identify symptoms based on given audio description. We proposed a framework known as TelMedAI which is designed to recognize patient speech to comprehend disease symptoms besides converting the speech text into desired language. The framework is useful for realizing a telemedicine system. Speech to Speech (STS) module takes the patient's audio content into English audio. STS module exploits the Bi-LSTM model with an encoder, decoder and attention mechanism for translation. Then Google Speech API is used to convert English audio into English text. Then the framework exploits Natural Language Processing (NLP) to improve the quality of text. Afterwards, the disease and symptoms miner module eventually recognizes a list of diseases and corresponding symptoms. We proposed an algorithm known as Learning based Disease and Symptom Recognition from Patient Speech (LbDSRPS). This algorithm has the functionality to develop TelMedAI which helps doctors in telemedicine. Our empirical study has revealed that TelMedAI takes technology-driven telemedicine research forward significantly. The highest accuracy achieved by the proposed framework is 68.13% which is much better than the baseline LSTM model used for voice translation.

MedPodGPT: A multilingual audio-augmented large language model for medical research and education.

The proliferation of medical podcasts has generated an extensive repository of audio content, rich in specialized terminology, diverse medical topics, and expert dialogues. Here we introduce a computational framework designed to enhance large language models (LLMs) by leveraging the informational content of publicly accessible medical podcast data. This dataset, comprising over 4, 300 hours of audio content, was transcribed to generate over 39 million text tokens. Our model, MedPodGPT, integrates the varied di-alogue found in medical podcasts to improve understanding of natural language nuances, cultural contexts, and medical knowledge. Evaluated across multiple benchmarks, MedPodGPT demonstrated an average improvement of 2.31% over standard open-source benchmarks and showcased an improvement of 2.58% in its zero-shot multilingual transfer ability, effectively generalizing to different linguistic contexts. By harnessing the untapped potential of podcast content, MedPodGPT advances natural language processing, offering enhanced capabilities for various applications in medical research and education.

Detecting Deep Fake Voice using Machine Learning

With the rapid advancement of deep learning techniques, the creation of synthetic media, particularly deep fake voices, has become increasingly sophisticated and accessible. This poses significant challenges in maintaining trust and authenticity in audio-based content. In response, this project proposes a machine learning-based approach to detect deep fake voices. The project begins by curating a diverse dataset consisting of genuine and deep fake voice samples, covering various demographics, accents, and emotional expressions. Pre-processing techniques are applied to clean and standardize the audio data, followed by feature extraction to capture relevant characteristics of the voice signals. For model development, a Convolutional Neural Network (CNN) architecture augmented with recurrent layers is employed, leveraging its ability to learn spatial and temporal features from the spectrogram representations of the audio. The model is trained on the prepared dataset using categorical cross-entropy loss and optimized through backpropagation.Evaluation of the trained model is conducted on a separate test set, measuring performance metrics such as accuracy, precision, recall, and F1-score. Post- processing methods, including thresholding and smoothing, are applied to refine the model’s predictions and enhance robustness. The proposed approach offers a promising framework for detecting deep fake voices in audio content, contributing to the ongoing efforts to combat the spread of misinformation and preserve the integrity of digital media. However, ongoing research and collaboration across disciplines are essential to address emerging challenges and ensure the responsible deployment of deep fake detection technologies.

Faculty of Management, Sremski Karlovci, “Union-Nikola Tesla“ University

Re-defining important changes in the music and film restoration industry, exploring the possibility to: 1. Extract individually chosen sounds from the final movie track in order to re-balance (re-mix) its audio content. 2. Extract chosen sounds from the original audio recording and re-mix, with separated audio stems. The Results section starts by describing the process of creating a completely new mixdown made from AI-generated audio stems (previously extracted from the original mono recordings) joining it with restored original 16 mm video footage, resulting in a full eight-hour movie, separated into three episodes, each with a duration between two and three hours, covering about one week each, of the 21 days of studio time. The material was recorded during the filming of The Beatles’ “Get Back” sessions at Twickenham Studios, London, in January 1969, and chronicles the making of “Let It Be” and the band’s famed final “Rooftop Concert.” The second episode describes the audio/video production of the last-ever song by The Beatles (“Now and Then”) literally made from scratch, as the core of the project was only a poor-quality analog recording made by John Lennon, singing and playing piano to an old, mono, compact cassette recorder.

The Discipline of Female Bodiness in TikTok Visual Music

This article aims to examine the TikTok music phenomenon among generation Z. The author uses a qualitative case study approach using textual analysis and critical paradigms. The object of study described is the music content published by Idgitaf and Mahalini on their official TikTok channel. There are 2 main points as foundings: the construction of a participatory culture carried out through the Duet Challenge feature and new visual music standards through the Filter Challenge promoted by TikTok. This participatory culture normalizes the standards for duets or performing songs today that can be done by professional musicians with fans/ordinary TikTok users (amateur singers). The existence of TikTok's visual features also emphasizes that TikTok's current standardization of digital music forms must emphasize the quality and beauty of its visuals compared to its audio content.

Neural Networks within Generative AI: A Review from a Marketing Research Perspective

Focusing on the role of neural networks within Generative Artificial Intelligence, this paper reviews their operating principles, recent developments, and implications for research in the marketing discipline. Special emphasis is placed on generative networks that produce synthetic data with a high degree of similarity to original data – such as realistic images, textual, or audio content – thereby expanding the potential for applied marketing practices. Besides significant opportunities opened up by Generative Artificial Intelligence, the review also elaborates on challenges, including technical obstacles like achieving an adequate Nash equilibrium between competing network components, and application-related considerations such as finding reliable metrics to evaluate their business performance. Broader implications are also analysed from recent publications, including ethical dilemmas related to data authenticity. As a result, this study provides a distinct perspective within the intersecting fields of marketing and computer science in the discourse surrounding Generative Artificial Intelligence, underscoring the research innovations arising from neural network models.

AI Integrity Solutions for Deepfake Identification and Prevention

The increasing complexity of deepfake technology has sparked significant worries over individual privacy, the spread of false information, and deficiencies in cybersecurity. Deepfakes have the ability to effectively modify audio and visual content, resulting in a growing challenge to differentiate between real and fake content. To address this critical challenge, the study is conducting a survey to reveal a broad range of perspectives on the familiarity, encounters, and concerns related to deepfake technology. In addition, the study evaluates the effectiveness of current strategies in addressing the spread of deepfake material and proposes future approaches for improving the integrity of AI. The survey was delivered digitally, and responses were examined to provide an in- depth analysis of the latest techniques and difficulties in the context of deepfake detection. The outcomes demonstrate a range of perspectives on understanding deepfakes with an explicit agreement on the importance of risk presented by harmful deepfake applications. Although the participants showed an understanding of the available interventions, they also identified considerable challenges and the need for improved awareness, robust detection tools, and ethical standards in AI development to address the challenges posed by deepfakes with the present detection method. Implementing AI ethical guidelines to avoid deepfakes has a significant and beneficial effect on several industries by providing protection against their harmful effects. By fostering AI ethical guidelines, these policies are able to foster societal trust, mitigate risk, and cultivate a more robust digital environment.