Results found: 2

Search results

Sort By:

Limit search:

Audio Stream Analysis for Deep Fake Threat Identification

100%

Jędrasiak K.

Civitas et Lex

2024

vol. 41

issue 1

21-35

This article introduces a novel approach for the identification of deep fake threats within audio streams, specifically targeting the detection of synthetic speech generated by text-to-speech (TTS) algorithms. At the heart of this system are two critical components: the Vocal Emotion Analysis (VEA) Network, which captures the emotional nuances expressed within speech, and the Supervised Classifier for Deepfake Detection, which utilizes the emotional features extracted by the VEA to distinguish between authentic and fabricated audio tracks. The system capitalizes on the nuanced deficit of deepfake algorithms in replicating the emotional complexity inherent in human speech, thus providing a semantic layer of analysis that enhances the detection process. The robustness of the proposed methodology has been rigorously evaluated across a variety of datasets, ensuring its efficacy is not confined to controlled conditions but extends to realistic and challenging environments. This was achieved through the use of data augmentation techniques, including the introduction of additive white noise, which serves to mimic the variabilities encountered in real-world audio processing. The results have shown that the system's performance is not only consistent across different datasets but also maintains high accuracy in the presence of background noise, particularly when trained with noise-augmented datasets. By leveraging emotional content as a distinctive feature and applying sophisticated machine learning techniques, it presents a robust framework for safeguarding against the manipulation of audio content. This methodological contribution is poised to enhance the integrity of digital communications in an era where synthetic media is proliferating at an unprecedented rate.

Algorithm for automatic analysis of communication skills of a participant in a public speaking VR exercise

51%

Bereska D., Jędrasiak K., Wolański R. M.

Ochrona ludności i dziedzictwa kulturowego

2024

vol. 5

81-109

Niniejsze badanie opracowuje i weryfikuje algorytm automatycznej analizy umiejętności komunikacyjnych w wystąpieniach publicznych w środowisku wirtualnej rzeczywistości (VR), rozwiązując ograniczenia tradycyjnych metod szkoleniowych. Badanie wprowadza algorytm do analizy kompleksowych danych, w tym wzorców mowy, wskazówek niewerbalnych i dostosowanych informacji zwrotnych, w szczególności dla użytkowników niepełnosprawnych. Metoda obejmuje wiele kroków: przetwarzanie danych wejściowych, analizę gestów, ruchów przestrzennych, tonu i barwy głosu oraz tempa mowy. Algorytm został przetestowany w oparciu o oceny ekspertów przy użyciu zestawu danych 20 nagrań wystąpień publicznych VR. To podejście oparte na VR zapewnia immersyjne, adaptacyjne środowisko szkoleniowe, kluczowe dla dziedzin o wysokiej stawce, takich jak służby ratunkowe, w których skuteczna komunikacja może mieć wpływ na wyniki w sytuacjach zagrażających życiu.

This research develops and validates an algorithm for automated analysis of communication skills in public speaking within a Virtual Reality (VR) environment, addressing limitations in traditional training methods. The study introduces an algorithm to analyze comprehensive data, including speech patterns, non-verbal cues, and tailored feedback, particularly for users with disabilities. The method involves multiple steps: processing input data, analyzing gestures, spatial movements, voice tone and timbre, and speech rate. The algorithm was tested against expert evaluations using a dataset of 20 VR public speaking recordings. This VR-based approach provides an immersive, adaptive training environment, crucial for high-stakes fields such as emergency services, where effective communication can impact outcomes in lifethreatening situations.

Refine search results

1 Civitas et Lex

1 Ochrona ludności i dziedzictwa kulturowego

2 Jędrasiak K.

1 Bereska D.

1 Wolański R. M.

2 2024

Search results

Audio Stream Analysis for Deep Fake Threat Identification

Algorithm for automatic analysis of communication skills of a participant in a public speaking VR exercise