Search Results (3)

Urban acoustic environments and performing-arts spaces strongly influence speech communication quality, acoustic comfort, and public wellbeing, particularly in noise-exposed shared environments such as transport hubs, campuses, healthcare spaces, public service facilities, music-education settings, and rehearsal or performance-related spaces. To address speech–noise separation in low signal-to-noise ratio and acoustically complex scenarios, this study proposes a lightweight two-stage deep learning framework termed LCGRU–Wave-SkipConvNet. In the preprocessing stage, a Lightweight Convolutional Gated Recurrent Unit (LCGRU) model is employed to achieve preliminary separation of target speech and background noise by capturing both spatial and temporal acoustic features. In the post-processing stage, a Wave-SkipConvNet model is introduced to further suppress residual noise and enhance speech quality. Experimental results demonstrate that the proposed framework achieves superior performance under different signal-to-noise ratios, sound-source angles, and target angle errors. For example, in the preprocessing stage, the LCGRU model achieved a perceptual evaluation of speech quality (PESQ) score of 2.64 at source angles between 0° and 30°, outperforming the convolutional neural network-long short-term memory (CNN-LSTM) model by 1.17. In the post-processing stage, the Wave-SkipConvNet model achieved higher short-time objective intelligibility (STOI) and segmental signal-to-noise ratio (segSNR) values than the comparison models under different SNR conditions. The proposed framework provides an effective and deployment-oriented AI solution for improving speech accessibility and acoustic comfort in urban acoustic environments and performing-arts spaces. Beyond speech enhancement, it offers practical potential for supporting healthier, more inclusive, and more equitable acoustic environments in noise-sensitive public and educational spaces. It should be noted that this study focuses on the objective acoustic environment and signal-level speech enhancement, rather than subjective soundscape perception, musical perception, or human perceptual evaluation. Full article

This study investigated how test room acoustic conditions relate to listening comprehension performance in a high-stakes English as a foreign language (EFL) assessment context. Using score data (n = 2532) from five TOEFL ITP test sessions conducted between 2021 and 2025 at a private university in Chiba, Japan, we compared performance across three lecture halls with documented differences in reverberation time (RT) and Speech Transmission Index (STI). Each listening score was linked to an approximated seat-based STI value, while grammar/reading scores were used to account for baseline proficiency. Linear mixed-effects modeling analyses indicated that examinees in the least favorable acoustic environment (RT_0.5–2kHz 1.51 s, STI 0.60) obtained lower listening scores than those in rooms with shorter RT (0.93 s, 0.79 s) and higher STI (0.69, 0.67), respectively. Subgroup analyses revealed a significant effect at the CEFR-J B1.1 level, though the room and B1.1 effects showed modest estimated marginal mean differences (EMM_Diff) roughly corresponding to 2–3 points on the total scale. Seat-based STI analyses also showed significant EMM_Diff, with approximately 3–7 total score point differences observed between categories F (0.52–0.55) and ≥D (≥0.60). While the dataset was limited to one institution and the sample distribution limited generalizability of the findings, the study offers empirical findings that can inform future research and discussions on equitable listening assessment practices. Full article

Background/Objectives: Despite its significant impact on learning, classroom acoustics and students’ hearing difficulties are often overlooked compared with more visible issues like lighting. Hearing loss—frequently underestimated and invisible—affects both students and teachers, potentially leading to fatigue, reduced participation, and academic challenges. The A.BA.CO. project in Italy was developed to address these issues by promoting improved classroom design, technological solutions, and better auditory communication accessibility in schools. Objective: This article presents the A.BA.CO. project, offering context and an overview of the preliminary analyses conducted by its multidisciplinary team. The goal is to share insights and propose organizational frameworks, technical solutions, and best practices concerning the hearing, communication, and auditory learning challenges experienced by students with hearing impairments. Results: The A.BA.CO. team’s analyses identified key barriers to inclusion for students with (or without) hearing impairments, such as poor classroom acoustics, excessive noise, and suboptimal seating arrangements. The project underscores the importance of improved acoustic environments and the integration of assistive technologies, including speech-to-text systems. The findings highlight the need for interdisciplinary collaboration to design accessible and inclusive educational settings for all learners. Conclusions: Embedding educational audiology within school systems—alongside enhancements in classroom acoustics and the use of assistive technologies and other technological solutions—is essential to ensure that all students, regardless of hearing ability, have equitable access to learning and full participation in educational life. Full article