China
Africa
Explore chapters and articles related to this topic
Objective Quality and Intelligibility Measures
Published in Philipos C. Loizou, Speech Enhancement, 2013
The speech-transmission index (STI), developed by Houtgast and Steeneken in the early 1970s [97], is based on an acoustic transmissions or systems approach. More precisely, it is based on the concept of modulation transfer function (MTF) which has been used as a design tool for auditorium acoustics. The MTF has been used as a measure for assessing the effects of a room (e.g., reverberation, etc.) on SI. For instance, given the volume (size) of the room, the reverberation time, ambient noise level, the talker’s vocal level, and talker-to-listener distance, one can derive analytically the MTF function and use that to derive a SII, called the speech transmission index [98]. As will be shown in the next sections, this index borrows some of the principles used in the AI but uses an indirect method for deriving the SNR in each band [98–100].
Studies on Acoustic Properties of Open-Plan Office Rooms
Published in Dariusz Pleban, Occupational Noise and Workplace Acoustics, 2020
An important factor affecting concentration of the mind among workers is keeping trace (consciously or not) of conversations carried out by other individuals present in the same room. Therefore, in an open-plan office room it is necessary to see to it that the speech intelligibility is as low as possible. The related acoustical quantity is called the speech transmission index and denoted STI. Table 7.2 lists the relationship between values of the speech transmission index STI and the speech intelligibility determined subjectively according to EN 60268-16:2011 and EN ISO 9921:2003.
Machine Learning in Acoustic DSP
Published in Francis F. Li, Trevor J. Cox, Digital Signal Processing in Audio and Acoustical Engineering, 2019
The use of machine audition of speech or music to determine acoustic parameters of the channel through which the speech and music have been transmitted has been studied, leading to the development of a number of new algorithms and methods. Reverberation time (RT) is the most common acoustic parameter used for the assessment of room acoustics. Early decay time (EDT) was reported to better correlate with human perception of the reverberation effect and, hence, has become an ever-popular acoustic parameter. Speech transmission index (STI) is an IEC standardised objective parameter for speech intelligibility assessment in a space. These parameters are considered in this chapter. Readers who are not familiar with room acoustics may find Kuttruff (2000) a wonderful resource to learn more about these parameters. Short-time RMS values of received discrete speech utterances, in particular pronounced digits, can be used to determine RT and EDT by statistical machine learning with an ANN (Cox, Li & Darlington, 2001a; Cox & Li, 2001b). STI can be accurately determined from received running speech signals with an envelope spectrum estimator and ANNs (Li & Cox, 2001; Li & Cox, 2003). If the background noise level is reasonably low, RT and EDT can be obtained from running speech in a similar way. Speech stimuli have a limited frequency range and, therefore, can only be used to determine acoustic parameters from 250 Hz to 4 kHz octave bands effectively. Music is considered to solve the problem, but the uneven spectra of music signals mitigate the accuracy. A note-matching filter bank can be used to circumvent the problem (Kendrick et al., 2007; Kendrick et al., 2008). Envelope spectra are a relatively stable feature of anechoic speech. With the aid of eigenvector as a feature space, blind estimation of STI and RT is possible (Li & Cox, 2007).
Roof Replacement of a Heritage Building Using Transparent Solutions: Room Acoustic Performance Comparison
Published in International Journal of Architectural Heritage, 2022
Macieira Mónica, Paulo Mendonça, João Miranda Guedes, António P. O. Carvalho
The STI (Speech Transmission Index) is a measure of speech transmission quality. It is one of the criteria used to characterize word perception. It is a dimensionless parameter that takes values between 0 and 1; the closer the value comes to 1, the better is the intelligibility of the spoken word. Table 4 shows the STI results obtained for each solution in an S01/receiver combination. They are scored by a colour scale: the darker the cell is, the better is the result. Except for GR+RMCr configuration, that presents a null score, all solutions are scored as poor (according to the subjective intelligibility scale of the normalized word (Patrício 2010)). The maximum STI mean value (0.41) is attributed to the WR+RMCr configuration, showing that this configuration is slightly more suitable for carrying out activities with word emission than the other ones. These STI results are indicative of poor speech intelligibility. The background noise level used in the modelling was of 75 dB and it was experimentally obtained in situ.
Characterizing the Indoor Acoustical Climate of the Religious and Secular Rock-Cut Structures of Cappadocia
Published in International Journal of Architectural Heritage, 2023
Ali Haider Adeeb, Zühre Sü-Gül, Ayşe Belgin Henry
Speech intelligibility can also be assessed by means of the Speech Transmission Index (STI), which is a metric that denotes the quality of speech while considering the loss of speech articulation caused by reverberation (Steeneken and Houtgast 1980). Figure 20 shows both male and female STIs for all the spaces studied. The differences between the average male and female STI values are considerably small in every rock-cut structure. The lowest STI value is 0.57 (STI female, Hallaç Church), whereas the highest value is 0.77 (STI male, Avanos). STI values between 0.60 and 0.75 indicate good speech intelligibility, while any value higher than 0.75 is considered excellent (Beranek 1988). Therefore, the spaces under study are proper for speech intelligibility.
Acoustical Footprint of the Traditional Turkish Baths in Historic Settings
Published in International Journal of Architectural Heritage, 2023
Zeynep Bora Özyurt, Zühre Sü Gül
Regarding Ottoman baths, live musical activities with instruments and chorus are a big part of the bathing tradition as well as verbal communication. Hence, the evaluation of STI, D50, and C80 parameters can provide essential information on the acoustic environment. Speech Transmission Index (STI) is a parameter to measure the quality of the intelligibility of speech (Beranek 1988). Table 5displays average female and male STI results of measured caldaria. No considerable difference is obtained between female and male STI values.