Colin Novak

In-vehicle speech intelligibility for different driving conditions using the Speech Transmission Index

The understanding of communications between passengers in an automobile as well as vehicle information systems to the driver is critical. A comprehensive study was conducted to evaluate speech intelligibility inside a vehicle for person to person communication using the Speech Transmission Index (STI). This metric provides a measure of the understanding or intelligibility of speech and has a value which varies from 0 for completely unintelligible to 1 for perfect intelligibility. For this study, several vehicle operating conditions, road surfaces and talker and listener configurations were compared in terms of the STI metric to gain a better understanding of these influences on in-vehicle person to person communication. Individual contributions of background noise and interior vehicle acoustics to the STI were also investigated. The results of this study found that higher in-vehicle sound pressure levels from engine and road noise are not necessarily associated with lower speech intelligibility. A wide range of STI values considered to be significant from subjective descriptions of intelligibility provided in the IEC EN 60268-16:2003 standard corresponding to those values were observed. These observations were for particular vehicle operating conditions, various configurations of the talker and the listener locations inside the vehicle, as well as between the ears of the listener at any particular location inside the vehicle. It was found that the rate of change of STI varied in unsteady background noise due to variable speed operation, at any given speed. It is recommended that subjective testing in a simulated vehicle environment be pursued to investigate any correlation of results with the STI metric. Also recommended is to determine the significance of the findings in terms of human perception of speech intelligibility inside a vehicle, and further investigate the appropriateness of using the STI metric, which is one of the most comprehensive speech intelligibility metrics available for a complete assessment of factors influencing in-vehicle person to person communication. Lastly, the effects of reverberation on reducing speech intelligibility were found to be negligible.

The evaluation of speech intelligibility in a simulated driving environment using the hearing in noise test

The implementation of the hearing in noise test (HINT) was carried out using an NVH driving simulator in order to evaluate the speech intelligibility in a car between driver and passenger for a variety of driving speeds and the configurations of the talker and of the listener. The sentence reception threshold (sSRT) was determined for each of the various communication situations. When presented with the same listening task, the participants required on average an approximate 3 dB increase in sound pressure level of the HINT speech material while driving and listening compared to when just listening, for an equivalent speech intelligibility performance. A suggested improvement to the current state of art would be to develop a driving simulation for acoustic perception jury testing as described in this study. This would form the basis of a standard method for a more complete, accurate and repeatable evaluation of in-vehicle speech intelligibility.

In-vehicle application of common speech intelligibility metrics

The purpose of this investigation is to quantify the loss of speech intelligibility related to communication between passengers inside a vehicle at different vehicle operating conditions and road surfaces using common objective speech intelligibility metrics. The goal was to identify the most appropriate metric, if a single one exists, for use in automotive applications. The objective metrics include the articulation index (AI), the speech intelligibility index (SII) and the speech transmission index (STI). The SII method, utilising user-defined, measured, speech signal was found to be the best out of the three metrics for quantifying in-vehicle speech intelligibility. Since the effect of reverberation on the loss of speech intelligibility was negligible, this method resulted in a close correlation with the more measurement-intensive STI method. This potentially provides a reduction in measurement effort while preserving the accuracy of the results.

Experimental investigations of noise and vibration in electric machines

Noise and vibration in electric machines is a significant issue to take into account when high performance is wanted to be achieved. During operation, vibration frequencies which match the resonance frequencies of the structure are absolutely to be avoided, because of their negative effect on the machine performance and the risk of mechanical damage. In this paper, a test procedure is described to check if this condition is achieved for a sample induction machine. In particular, two kinds of test are performed: a modal test, which gives information about the resonance frequencies, and a test with the machine operating at different speeds, in order to evaluate the frequency spectrum of the vibrations during working condition.

Determining binaural summation for stationary signals across the frequency spectrum

The concept of binaural summation has been widely accepted; however, the value of summation, and the manner in which it is applied, is still under a great deal of debate. This research is an investigation of the binaural summation mechanisms through the use of a loudness comparison experiment for pure tones. Preliminary results have concluded that for pure tones presented at 40 dB the amount of summation increases with frequency from approximately 2 dB at 40 Hz up to 8 dB at 10 kHz. Due to the low hearing threshold at 40 dB, the experiment was modified such that the hearing threshold of each individual is collected in order to presented each signal at a fixed amplitude above the threshold of hearing for each frequency (i.e., a sensation level of 20 dB(SL)). However, similar outcomes have been found using this approach, thus supporting the conclusions that binaural summation does exist and is found to increase at defined increments with increasing frequency for pure tones. The eventual outcome of this work...

Comparative study of a low frequency emergency siren to traditional siren technology

This study measures and compares the acoustic characteristics of a traditional electronic emergency siren and an innovative low frequency Rumbler siren technology. The Rumbler siren’s low frequency emissions can travel further and have a greater ability to penetrate and induce structure-bone excitation in nearby vehicle cabins compared to other siren technology. The result is a better ability to warn both nearby vehicles and pedestrians thus lessening the potential of emergency vehicle collisions. The siren technologies were evaluated using three measurement scenarios to determine acoustic localization characteristics, drive-by effectiveness, and the ability to overcome the problem shadowing phenomenon typical at congested intersections. A comparative analysis of the acoustical characteristics of the standalone siren and the addition of The Rumbler system provided insight in regards to the relative effectiveness of each siren mode. The siren system equipped with The Rumbler technology gave a noticeable in...

Study of acoustic localization of an emergency siren

Having the ability to adequately detect the direction of an approaching emergency siren is critical to the effectiveness of the emergency system. Having this ability allows both pedestrians and drivers of nearby vehicles to more quickly and safely react to an approaching emergency vehicle. This study considers a typical electronic siren system that is currently being used by the Windsor Fire & Rescue Services Department. This siren has two fundamental settings: the standard siren signal and the air horn mode, which is typically used when the emergency vehicle is approaching roadway intersections, as these pose the most danger to occupants of both emergency vehicle and general public. Siren and air horn signals were recorded at specific distances from the driver’s position at 45° radial increments. From these, the recorded signals and sound pressure levels measured inside the cabin of the vehicle at the various approach angles were used to prepare a subjective jury evaluation to determine the localization ...

Study of a low frequency emergency siren in comparison to traditional siren technology

This study measures and compares the acoustic characteristics of a traditional electronic emergency siren and an innovative low frequency Rumbler siren technology. The Rumbler siren’s low frequency emissions can travel further and have a greater ability to penetrate and induce structure-bone excitation in nearby vehicle cabins compared to other siren technology. The result is a better ability to warn both nearby vehicles and pedestrians thus lessening the potential of emergency vehicle collisions. The siren technologies were evaluated using three measurement scenarios to determine acoustic localization characteristics, drive-by effectiveness and the ability to overcome the problem shadowing phenomenon typical at congested intersections. A comparative analysis of the acoustical characteristics of the standalone siren and the addition of The Rumbler system provided insight in regard to the relative effectiveness of each siren mode. The siren system equipped with The Rumbler technology gave a noticeable increase in sound pressure level for each of the three test scenarios. Through analysis of the measured data the overall performance of the emergency siren system with the Rumbler technology was shown to be a more effective emergency notification device compared to the standalone electronic siren.This study measures and compares the acoustic characteristics of a traditional electronic emergency siren and an innovative low frequency Rumbler siren technology. The Rumbler siren’s low frequency emissions can travel further and have a greater ability to penetrate and induce structure-bone excitation in nearby vehicle cabins compared to other siren technology. The result is a better ability to warn both nearby vehicles and pedestrians thus lessening the potential of emergency vehicle collisions. The siren technologies were evaluated using three measurement scenarios to determine acoustic localization characteristics, drive-by effectiveness and the ability to overcome the problem shadowing phenomenon typical at congested intersections. A comparative analysis of the acoustical characteristics of the standalone siren and the addition of The Rumbler system provided insight in regard to the relative effectiveness of each siren mode. The siren system equipped with The Rumbler technology gave a noticeable incr...

Acoustic localization of an electronic emergency siren

Having the ability to adequately detect the direction of an approaching emergency siren is critical to the effectiveness of the emergency system. Having this ability allows both pedestrians and drivers of nearby vehicles to more quickly and safely react to an approaching emergency vehicle. This study considers a typical electronic siren system that is currently being used by the Windsor Fire & Rescue Services Department. This siren has two fundamental settings; the standard siren signal and the air horn mode, which is typically used when the emergency vehicle is approaching roadway intersections, as these pose the most danger to occupants of both emergency vehicle and general public. Siren and air horn signals were recorded at specific distances from the drivers position at 45° radial increments. From these, the recorded signals and sound pressure levels measured inside the cabin of the vehicle at the various approach angles was used to prepare a subjective jury evaluation to determine the localization c...

Validating a binaural head for use in jury testing

A test procedure for use in loudness perception tests must be created to completely describes a phenomenon while at the same time minimizing jury listening fatigue. One contributor to this fatigue is the amount of time necessary for the test subject to experience all the required signals. Head and torso simulators have been used for years as a means to reliably quantify the acoustic performance of a product while avoiding the influence of listener bias and fatigue. This procedure not only controls the test parameters but also removes any human error that may occur. The purpose of this investigation is to qualify a head and torso simulator for use in loudness investigations. The objective of this experiment is to correlate the results from using this equipment to human subject results for high resolution experiments on directionality of loudness.