Go to the NASA Homepage
 
Advanced Controls and Displays Left-Side Header Image
Research Sidebar Header
Advanced Controls and Displays Image Collage
Research header
Multiple speech communication systems
Principal Investigators
Durand R. Begault, Elizabeth M. Wenzel

Problem
Certain radio communication personnel are challenged with multiple channel listening requirements, including emergency (911) telephone operators, military helicopter pilots, and the space shuttle launch personnel at NASA's John F. Kennedy Space Center (KSC). For example, the current practice at KSC is to use single ear piece headsets, connected to a radio device capable of monitoring either four or eight separate channels. This situation is similar to the working conditions of police and fire service 911 operators, who use a combination of single ear piece headsets, telephones, and loudspeakers. The situation parallels teleconferencing with multiple participants. It is significantly more difficult to separate out multiple streams of auditory information with only one ear compared to two-ear listening. The level of a desired channel must be significantly increased relative to other channels, enhancing the possibility of fatigue or hearing loss.

Approach and Objectives
Our objective is safer and more intelligible radio and teleconferencing communications. Presentation of spatially separated speech sources has long known to be superior to monaural listening. However, previous methods used for spatially separating signals in previous systems (usually simple overall phase or intensity differences at the two ears) are do not yield as good an advantage as 3-D audio techniques. Simple forms of spatialization are not representative of our "natural" everyday binaural advantage, primarily due to the lack of the HRTF spectral modification, resulting in inside-the-head locatedness. Listening over headphones with a 3-D audio system, one has a spatial sense of each channel originating from a unique position outside the head; i.e., as if four people were standing about you, speaking from different directions. As opposed to monotic (one ear) listening, the typical situation in communications operations, binaural listening allows a listener to use head-shadow and binaural interaction advantages simultaneously. This allows about a 6 dB improvement in intelligibility, compared to one ear listening.

Our approach was to evaluate an HRTF-based intelligibility system to determine the optimal placement for maximal intelligibility, using virtual acoustic (3-D audio) techniques. and then compare results to normal, two-ear hearing. Once this was verified, we designed and fabricated several prototype devices for both NASA KSC and JSC, called the Ames Spatial Auditory Display (ASAD; Figure 2.1).

Accomplishments
Significant progress was made towards addressing the following hypotheses: (1) Does the binaural intelligibility advantage known to exist in normal listening transfer successfully (equivalently) to a virtual acoustic display? (2) Does the limited frequency bandwidth of communication systems significantly affect the virtual acoustic intelligibility advantage of such a system?

To test these hypotheses, we used a standard method (ANSI) for gauging the intelligibility of speech over communication systems. The thresholds for speech material in the presence of noise was determined as a function of specific noise and signal position. 8- 10 subjects were used in each of the experiments associated with this project; all were screened before the experiment for normal hearing. For communication call signs, results for 8 subjects showed a maximum intelligibility improvement of about 6-7 dB when signals were spatialized to 60 or 90 degrees azimuth. In that experiment, the noise masker consisted of speech babble. For modified rhyme test word lists, an advantage of 5 dB was found against unmodulated speech spectrum noise. For limited frequency bandwidths, the binaural advantage was as high as 7 dB. Within a virtual acoustic display, intelligibility improves as a function of the interaural time delay inherent in the HRTFs used.

Significant progress was also made in terms of technology transfer. The Ames Spatial Auditory Display (U.S. Patent 5,438,623) places four different communication channels at fixed virtual auditory positions about the listener. The ASAD is designed to be as "user-friendly" as possible. A user therefore only needs to power the device up, and adjust the volume. This is in contrast to other "general purpose" spatial auditory displays that require a computer or a front panel interface for operation. Input channels to the spatial auditory display can be assigned to any position because the ASAD uses four removable EPROMs (electrical erasable programmable memory), each corresponding to a particular target position. The EPROMs themselves can contain spatial cues for any given position, measured at the user's ear or from a modeled or "averaged" ear.

One unexpected advantage found with the ASAD is that it allows a more hands-free operation of the host communication device. Normally communication personnel bring the volume up for a desired channel on the front panel of the radio in order to hear that channel over undesired channels. With the ASAD, one can simply direct attention to the desired stream. But a further advantage is that overall intensity level at the headset can remain lower for an equivalent level of intelligibility, an important consideration in light of the stress and auditory fatigue that can occur when one is "on headset" for an eight-hour shift. Lower listening levels over headphones could possibly alleviate the need to raise the intensity of one's own voice (known as the Lombard Reflex), reducing overall fatigue, and thereby enhancing both occupational and operational safety and efficiency of multiple-communication channel contexts.

Future plans
Future plans involve determining the optimal placement for more complex combinations of communication streams, and for completing the transfer of the technology to the private sector via a licensing agreement with the private sector. We also plan to study the advantage of including personalized HRTFs within the system.

Key references
Begault, D. R. (1993): Call sign intelligibility improvement using a spatial auditory display (Technical memorandum No. 104014). NASA Ames Research Center.
Begault, D. R., and Erbe, T. (1994). Multichannel spatial auditory display for speech communications. Journal of the Audio Engineering Society 42, 819- 826.
Begault, D. R. (1995). Virtual acoustic displays for teleconferencing: Intelligibility advantage for "telephone grade" audio. Audio Engineering Society 98th Convention preprint 4008. City: New York: Audio Engineering Society.
Begault, D. R. (1995): Multi-Channel Spatialization System for Audio Signals (United States Patent 5,438,623). Washington, DC: Commissioner of Patents and Trademarks.
Click to view - Figure 2.1. The Ames Spatial Auditory Display (ASAD)
Figure 2.1.
The Ames Spatial Auditory Display (ASAD).
Go to the First Gov Homepage
Go to the NASA - National Aeronautics and Space Administration Homepage
Curator: Phil So
NASA Official: Brent Beutter
Last Updated: August 15, 2019