Browsing by Subject "Speech intelligibility"

Now showing 1 - 7 of 7

Open Access
The acoustical characteristics of the Kocatepe Mosque in Ankara, Turkey
(Taylor & Francis, 2008) Sü, Z.; Yilmazer, S.
This research investigates the acoustical characteristics of mosques of the classical Ottoman period and the contemporary period. Kocatepe Mosque, the major mosque in Ankara, Turkey, as a case from the contemporary era is a unique combination of 16th century Ottoman aesthetics and 20th century technology. Although previous and recent mosques have been inspired by this combination, the use of a reinforced concrete dome without cavity resonators could cause long reverberation times especially in low frequencies. The purpose of this study was to investigate this condition by studying Kocatepe Mosque. The acoustical characteristics of the mosque were analyzed by computer simulation. Objective room-acoustic indicators including reverberation time (RT), early decay time (EDT), clarity (C80), definition (D50), lateral fraction (LF), speech transmission index (STI) and strength (G) are presented. The results show that the acoustical quality of Kocatepe Mosque is not optimal when it is empty, and closest to optimal conditions when fully occupied. The new techniques to overcome excessive low frequency attenuation caused by huge concrete shells of today's mosques are discussed in conclusion.
Open Access
A comparison study on sound defects of Bilkent Amphitheatre in terms of intelligibility of the speech
(2007) Yilmazer, S.
The 4000 seated multi-purpose hall Bilkent Amphitheater is a semi-closed amphitheater with a large volume of 90,000 m3.At the beginning, the amphitheater was designed only for open-air performances. Afterwards it was decided to be roofed with a tensile membrane in order to protect the spectators against atmospheric effects. However, the covered amphitheater was turned into a very resounding and a problematic place with a reverberation time even more than 6 s. when the hall is unoccupied. After that time, some acoustical interventions were implemented, but these have not been sufficient in decreasing the number of problematic areas of echoes and dead spots. In 2000, Akukon Oy Consulting Engineers, Finland, carried out the real-size measurements when the hall was unoccupied, in order to find out the values of different acoustical parameters. In 2004, the computer simulation of the hall for the unoccupied condition was made [2]. In order to test speech intelligibility in the hall, besides the computer simulation and real size measurements, in 2006, articulation testing (AI) was made by using normal hearing listeners in the hall. In this paper, comparison between the measurements and the experimental studies are presented. Results confirm that excessive values of reverberation time cause too much reverberant and liveliness throughout the hall, besides the lack of warmth subjectively. Clarity is undermined by excessive reverberate and the sound becomes blurred. Sound energy are not satisfactory in reflecting the sound sufficiently through the hall and the direct sound becomes insufficient at some locations as back and mid sides.
Open Access
Developing wideband acoustic absorbers in interior spaces: modelling, measurements and applications
(2024-01) Fasllija, Ela
This study aims to develop wideband acoustic absorbers in interior spaces for speech purposes, through the use of Micro-Perforated Panels (MPPs) arranged in different configurations. The research has involved theoretical modelling, measurements, and applications in interior spaces. The initial phase focused on modelling various MPP arrangements and predicting their sound absorption coefficients using the Electric Equivalent Circuit Model (ECM) under normal and oblique incidence conditions. Small samples with different geometric parameters were then manufactured and tested for their absorption coefficients at the Turkish Standards Institute, following the impedance tube method (ISO 10534-2:1998). The preliminary results showed a good agreement with the numerical predictions. Furthermore, the random incidence absorption coefficient of a parallel arranged MPP structure was measured in a Small-Scale Reverberation Room (SSRR) at Bilkent University to evaluate the proposed structure in a diffuse field. The results obtained by using ISO 354:2003 showed lower values than the predicted ones by a range of 0.1- 0.3. However, there was a better alignment between the ECM predictions and experimental results for 30-degree oblique incidence. Regarding interior applications, a classroom with poor acoustics was treated in ODEON 16 using the proposed material in the form of a wall panel. The effect of the developed material on objective room acoustics parameters was examined, revealing improvements in the Reverberation Time and Speech Intelligibility Index values. Lastly, the Turkish versions of Perceived Affective Qualities outlined in the ISO/TS 12913-2:2018 standard were developed to assess the subjective perception of the material following the soundscape approach.
Open Access
Effect of speech intelligibility on visual short-term memory performance
(2007) Kitapci, K.; Yılmazer, Semiha; Erkip F.
The aim of this study is to find out effects of speech intelligibility on visual short-term memory performance by testing 34 interior architecture students in three different sound environments. Those sound environments were composed of real office sounds and a speech sample that are mixed to achieve varying STI levels of 'continuous noise', 'masked speech' and 'speech' sound environments [2]. Test procedure had three steps: first questionnaire, computer based short-term memory test and second questionnaire. Results show that there is no significant difference of visual short-term memory performance between varying sound environments, however subjective results derived from questionnaires show that students were distracted most by 'speech' sound environment.
Open Access
An experimental study on speech interference in public leisure spaces
(Institute of Noise Control Engineering, 2010) Yilmazer, Semiha; Gezginer, Pelin Meriç
Physical comfort requirements of users in the non-acoustical spaces, leisure space, and usability of the space in long-term period should be taken into consideration in terms of sustainability of the space. The aim of this study is to investigate the speech interference of users' in an openplaned public leisure space. Bilkent Rollhouse was chosen as a leisure space because of its variance of activities in an open-plan like bowling, dart, billiards, play station and dining area. In this respect, a questionnaire was prepared to check the perception of sound of different user types in the area. The questionnaires were applied at the most crowded day of the week that is Friday and in between 18:00 to 21:00. The results of the survey were evaluated statistically. In addition to this, the equivalent continuous noise level, Leq, of the leisure space were taken in parallel to the questionnaire application to the users. Measuring of speech intelligibility via articulation index (AI) subjectively was the second phase of this study. Via application of the word list to the normal hearing users, the AI contours were prepared. Finally, objective and subjective measurements were analyzed in terms of speech interference during leisure activity in the space. As a leisure space, the relation between noise annoyance ratings and Leq did not show any significant difference in dart, billiards and play stations activities. However, in bowling activity and dining area, it was found that there is a significant relation between noise annoyance and Leq. This study reveals that users submit ambient noise of the space during leisure activity, they get in a common behaviour psychologically, they adjust their vocal noise that they create and finally these issues lead controlling of ambient noise by them. It means that users give up loudly speaking, they accept repeating or silence, focus on meal and they do not like getting in uncomfortable situation since their main activity is dining. For further study, the number of the samples in not only different leisure spaces, but also measurement points and sampling times should be increased to make clear hypothesis given above, since the study reveals completely different behaviour that is introduced in literature like cocktail party effect phenomenon.
Open Access
Investigating the potential of transparent parallel-arranged micro-perforated panels (MPPs) as sound absorbers in classrooms
(MDPI AG, 2023-01-13) Fasllija, Ela; Yılmazer, Semiha
Acoustic deficiencies due to lack of absorption in indoor spaces may sometime render significant buildings unfit for their purpose, especially the ones used as speech auditoria. This study investigates the potential of designing wideband acoustic absorbers composed of parallel arranged micro-perforated panels (MPPs), known as efficient absorbers that do not need any other fibrous/porous material to have a high absorptive performance. It aims to integrate architectural trends such as transparency and the use of raw materials with acoustical constraints to ensure optimal indoor acoustic conditions. It proposes a structure composed of four parallel-arranged MPPs, which have been theoretically modelled using the electrical Equivalent Circuit Model (ECM) and implemented on an acrylic prototype using recent techniques such as CNC machining tools. The resulting samples are experimentally analysed for their absorption efficiency through the ISO-10534-2 method in an impedance tube. The results show that the prediction model and the experimental data are in good agreement. Afterward, the investigation focuses on applying the most absorptive MPP structure in a classroom without acoustic treatment through numerical simulations in ODEON 16 Acoustics Software. When the proposed material is installed as a wall panel, the results show an improvement toward optimum values in Reverberation Time (RT30) and Speech Transmission Index (STI).
Open Access
Source and filter estimation for Throat-Microphone speech enhancement
(Institute of Electrical and Electronics Engineers Inc., 2016) Turan, M. A. T.; Erzin, E.
In this paper, we propose a new statistical enhancement system for throat microphone recordings through source and filter separation. Throat microphones (TM) are skin-attached piezoelectric sensors that can capture speech sound signals in the form of tissue vibrations. Due to their limited bandwidth, TM recorded speech suffers from intelligibility and naturalness. In this paper, we investigate learning phone-dependent Gaussian mixture model (GMM)-based statistical mappings using parallel recordings of acoustic microphone (AM) and TM for enhancement of the spectral envelope and excitation signals of the TM speech. The proposed mappings address the phone-dependent variability of tissue conduction with TM recordings. While the spectral envelope mapping estimates the line spectral frequency (LSF) representation of AM from TM recordings, the excitation mapping is constructed based on the spectral energy difference (SED) of AM and TM excitation signals. The excitation enhancement is modeled as an estimation of the SED features from the TM signal. The proposed enhancement system is evaluated using both objective and subjective tests. Objective evaluations are performed with the log-spectral distortion (LSD), the wideband perceptual evaluation of speech quality (PESQ) and mean-squared error (MSE) metrics. Subjective evaluations are performed with an A/B comparison test. Experimental results indicate that the proposed phone-dependent mappings exhibit enhancements over phone-independent mappings. Furthermore enhancement of the TM excitation through statistical mappings of the SED features introduces significant objective and subjective performance improvements to the enhancement of TM recordings. ©2015 IEEE.