Search

Military training and international operations require different nationalities to communicate in a common language, where there are potential challenges with non-native (L2) speech communication. An experiment of speech communication in military noise was conducted for co-located (face-to-face [F2F]) and distributed (using communication headsets) talker-listener pairs. Half of the twenty-four participants were monolingual English speakers (native group, NA) and the remaining half had obtained English fluency after the age of eight years (non-native group, NN). Two tests of speech understanding were used: the Modified Rhyme Test (MRT) and the Speech Perception in Noise test (SPIN). In the F2F condition, the participants wore a communication headset (earmuff) with the power off for occluded listening. Three levels of armoured vehicle noise were used, 55, 60 and 65 dBA, for speech-to-noise ratios ranging from −10 to +5 dB. In the radio condition, the pairs were separated by a visual barrier and used the communication headset for the tests in 80 dBA armoured vehicle noise. The results showed that the NN group had difficulty with the SPIN test in the radio and F2F conditions. This result was attributed to the open-response set of the SPIN. Headset occlusion likely contributed to the lower scores for the NN listeners in the F2F condition. There was a main effect of talker for the MRT in the F2F and radio conditions, and for the SPIN in the radio condition, suggesting that foreign accent reduced the intelligibility for both the NA and NN groups. The results were surprising considering the high L2 proficiency of the NN group. Training methods for improving L2 communication in operational settings should be further investigated.

In this paper, we have proposed a new algorithm considering commutation error and feedback effect to enhance the convergence rate and noise reduction efficiency of ANC controller. In order to improve noise reduction performance of the ANC headset with fixed-point DSP, we have proposed a new FxLMS algorithm, FxLMS CF, which considers the commutation error and feedback. Also, using a non-real-time simulation, we have decided the phase and amplitude compensation factors of anti-noise signal considering round-off and quantization error, nonlinear distortion and delay of analog device. We estimated the phase and amplitude compensation factors by simulation without using any special measuring devices or analysis devices, and reduced the broadband noise by 24 dB.

This paper reports on a study to assess the feasibility of creating an intuitive environmental sound monitoring system that can be used on-location and return meaningful measurements beyond the standard L Aeq . An iOS app was created using Machine Learning (ML) and Augmented Reality (AR) in conjunction with the Sennheiser AMBEO Smart Headset in order to test this. The app returns readings indicating the human, natural and mechanical sound content of the local acoustic scene, and implements four virtual sound objects which the user can place in the scene to observe their effect on the readings. Testing at various types of urban locations indicates that the app returns meaningful ratings for natural and mechanical sound, though the pattern of variation in the ratings for human sound is less clear. Adding the virtual objects largely has no significant effect aside from the car object, which significantly increases mechanical ratings. Results indicate that using ML to provide meaningful on-location sound monitoring is feasible, though the performance of the app developed could be improved given additional calibration.

Intercom headsets are mandatory communication apparatus in high noise environments (HNE). The headset selection in HNE, such as combat vehicles, is crucial for achieving the objectives of communication, as it serves the needs for both noise reduction and voice reproduction. Although military-grade intercom headsets are typically used under extreme environmental conditions, a standard performance evaluation method exists only for the earphone elements. In the present work we propose an integrated method for the assessment of the electroacoustic performance of HNE headsets in conditions of maximum reproduction level and high environmental noise, focusing on the voice communication quality. Objective methods, such as Automatic Speech Recognition (ASR), Perceptual Evaluation of Speech Quality (PESQ) and Speech Transmission Index (STI) are comparatively evaluated and their results are compared to subjective scores using Multiple Stimuli with Hidden Reference and Anchor (MUSHRA) in order to reveal the best fit metrics.

This paper presents the results of objective tests performed on 13 personal active noise reduction devices (earmuffs, headphones, headsets and insert earphones) divided into four groups based mainly on structure, using an acoustic test fixture (ATF). Each device was examined on its attenuation of broadband noise, overload response, internally generated noise, attenuation of impulse noise, and stability to movement. The results show a large range of responses between devices in terms of attenuation and overload, and highlight distinctive differences between the device groupings.

An experiment was conducted to determine the effect on speech understanding of an earplug and muff, worn in combination. Two groups of 12 normal-hearing subjects were tested individually in a semi-reverberant chamber. The measurements included free-field assessment of hearing thresholds from 0.25 to 8 kHz and consonant discrimination, with the ears unoccluded or fitted with E-A-R foam plugs, the Peltor Aviation communication headset, or the two devices in combination. The headset had active noise reduction (ANR) capability, and was tested without and with ANR operational. The speech was presented at a level of 85 dB SPL either in speech spectrum noise (Group 1) or cable swager noise (Group 2), presented with speech-to-noise ratios ranging from +5 to −10 dB. The two noise backgrounds differed in terms of both energy spectrum and temporal characteristics. The results showed that below 2 kHz, the devices worn in combination provided as much as 30 dB more sound attenuation than either device alone. The effect of the combination on consonant discrimination was a decrease of only 4%. The type of noise background and speech-to-noise ratio proved to be more important determinants of outcome.

The noise attenuation properties of aviation helmets and headsets are generally given only in commercial brochures, and attenuation comparisons between different types of helmets and headsets are seldom shown. Therefore a comparative study was carried out. Four military fighter helmets and two headset types were studied against pink noise in an anechoic chamber. In addition, the active noise reduction (ANR) system of one helmet and the deterioration effect of the noise attenuation of an Alpha aviation helmet in combination with sunglasses and a cushion cloth were studied. When the A-weighted noise attenuation values were compared, the noise attenuation value was 31 dB for the Silenta Super earmuffs, 27 dB for the Ampliguard earmuffs, 25 dB for the Alpha helmet with ANR, 22 dB for the Alpha helmet, 21 dB for the FFV 113B helmet, 16 dB (unadjusted) for the ACS helmet, and 10 dB for the ZSH-5 helmet. The sunglasses reduced the noise attenuation by 4 dB, and the cushion cloth only by 1–2 dB. ANR improved noise attenuation by 3–19 dB in the 63–250 Hz frequency range. We concluded that noise attenuation of greater than 20 dB is needed because modern fighters and helicopters often have a noise level higher than 100 dB in the cockpit. Therefore, the noise attenuation properties of aviation helmets should not be forgotten in helmet and headset design.

A method for measuring noise exposure from communication headsets is presented. It is based on an accurate acousto-mechanical model of the human head, and a filter to transform the data from the eardrum level to equivalent diffuse free-field values. The filter fits an existing integrating sound-level meter (Bruel and Kjaer type 2231), allowing measurements in real-time. The measurements were validated by comparing them with those obtained with a sound-level meter in the free field and with a miniature microphone. The method is accurate, objective, relatively simple, and works with any kind of communication headset. Measurements of noise exposure from communication headsets were taken at eight sites. The noise exposure level Lex,8(the equivalent sound level Leq normalized to 8 h) varied from 64 to 81 dBA in quiet office settings and from 77 to 88 dBA in noisy environments, with one measurement at 95 dBA.

The acoustic performance of five prototype in-ear hearing protectors was measured in the laboratory and in a military vehicle. Where a conventional or Active Noise Reduction (ANR) circum-aural headset is worn over a protective hood and/or respirator an acoustic leak occurs, degrading the hearing protection. In a noisy environment which contains a predominance of frequencies below 1 kHz and where it is essential to wear other headgear, a possible solution is offered by ANR in-ear hearing protection. The acoustic performance was measured using the in-built sense microphone. Noise at the ear was reduced from a sound pressure level of 112 dB(A) to below 90 dB(A) with the ANR ‘on’. The combined passive and active attenuation provided better protection than passive attenuation only. The acoustic performance of these prototypes was satisfactory, but the prototypes need to be easier to fit and more robust for military applications.