The study evaluated two speech recognition systems, Wav2vec2 and Whisper, for potential biases for Dutch speakers.Results obtained by evaluating on the JASMIN corpus revealed biases against non-native speakers, children, and the elderly,with (slightly) better performance for women. The study emphasizes the need for ASR systems to handle variations in speakingin order to reach equal performance among all users.
The diverse European landscape of climate consciousness is shaped by political values, financial constraints, and country-specific point of view. The aim of the study was to unravel age-specific ecological awareness, forms of engagement, and perceptions, contributing to a nuanced understanding of climate dynamics.Selected regions: Germany (Rheinisches Revier), the Netherlands (Amsterdam Metropolitan Area), and Poland (Upper Silesia/Metropolis GZM) present different states regarding recycling/Circular Economy principles, and different environments.The research design incorporates an inductive qualitative approach to investigate environmental awareness and attitudes toward ecologically friendly behaviors. Six FGIs (Focus Group Interviews) were conducted across three European regions, involving participants from diverse age groups (20–39 years and 40–60 years) in each region.The study shows that ecological awareness varies between countries and generations, reflecting distinctive environmental strategies shaped by cultural and developmental factors. Participants in each region and age group exhibit diverse levels of engagement in sustainable activities, and highlight issues such as the need for tailored strategies, concerns related to eco-labelling, greenwashing, and inadequate waste treatment, as well as information gaps.These variations in pro-environmental attitudes and behaviors across age groups and regions underscore the need for tailored strategies and regional policies. Transparency in waste management, eco-labelling, and sustainable transportation alternatives should be prioritized. Educational initiatives addressing information gaps, especially regarding lifestyle choices, are crucial. Collaboration and interdisciplinary approaches are essential for fostering positive change and a sustainable future across the European Union. Transparent communication, regulatory measures, and accessible eco-friendly options encourage widespread adoption of pro-environmental behaviors.
It is crucial that ASR systems can handle the wide range of variations in speech of speakers from different demographic groups, with different speaking styles, and of speakers with (dis)abilities. A potential quality-of-service harm arises when ASR systems do not perform equally well for everyone. ASR systems may exhibit bias against certain types of speech, such as non-native accents, different age groups and gender. In this study, we evaluate two widely-used neural network-based architectures: Wav2vec2 and Whisper on potential biases for Dutch speakers. We used the Dutch speech corpus JASMIN as a test set containing read and conversational speech in a human-machine interaction setting. The results reveal a significant bias against non-natives, children and elderly and some regional dialects. The ASR systems generally perform slightly better for women than for men.
MULTIFILE
