Audio Engineering Society

Convention Paper
Presented at the 136th Convention
2014 April 26–29 Berlin, Germany

This Convention paper was selected based on a submitted abstract and 750-word precis that have been peer reviewed
by at least two qualified anonymous reviewers. The complete manuscript was not peer reviewed. This convention
paper has been reproduced from the author’s advance manuscript without editing, corrections, or consideration by the
Review Board. The AES takes no responsibility for the contents. Additional papers may be obtained by sending request
and remittance to Audio Engineering Society, 60 East 42nd Street, New York, New York 10165-2520, USA; also see
www.aes.org. All rights reserved. Reproduction of this paper, or any portion thereof, is not permitted without direct
permission from the Journal of the Audio Engineering Society.

Ambidio: Sound Stage Width Extension for

Internal Laptop Loudspeakers
Tsai-Yi Wu1 , Agnieszka Roginska2 , and Ralph Glasgal3
1,2
Music Technology, New York University, New York, NY, 10012, USA
3
Ambiophonics Institute, Rockleigh, New Jersey, 07647, USA

Correspondence should be addressed to Tsai-Yi Wu ([email protected])

ABSTRACT
This paper introduces a sound stage width extension method for internal loudspeakers. Ambidio is a real-
time application that enhances a stereo sound file playing on a laptop in order to provide a more immersive
experience over built-in laptop loudspeakers. The method, based on Ambiophonics principles, is relatively
robust to a listener’s head position, and requires no measured/synthesized HRTFs. The key novelty of the
approach is the pre/post-processing algorithm that dynamically tracks the image spread and modifies it to
fit the hardware setting in real-time. Two detailed evaluations are provided to assess the robustness of the
proposed method. Experimental results show that the average perceived stage width of Ambidio is 176° using
internal speakers, while keeping a relatively flat frequency response and a higher user preference rating.

1. INTRODUCTION speakers is limited by the tight space constraints,

Mobile devices with stereo speakers such as lap- and becomes even tinnier when it goes across the
tops, tablets are increasingly popular. According to keyboard. This results in a narrow and unrealis-
Google’s research people spend 4.4 hours on average tic stereo image that makes all sound perceived as
in front of screens each day during leisure time [14] coming from one direction, and appear to be more
and they mostly use their laptop for entertainment. monophonic. There is a need to improve the sound
[19] Even though audio is essential to most popular quality of built-in laptop loudspeaker sound to ac-
laptop entertainment, e.g. games, movies, and mu- company the rich graphics users normally get from
sic, the performance of laptop internal speakers is the screen, for more immersive experiences.
not yet satisfactory. [9] Indeed, the sound of laptop
Wu et al. Ambidio: Sound Stage Width Extension for Internal Laptop Loudspeakers

The perceived width can be widened by applying tual speakers located further apart than the real ones
spatial enhancement techniques. [24] Achieving spa- as in normal Blumlein stereo. The standard was to
tial enhancement passively during playback is ideal use generalized HRTFs to represent the angles of the
for laptop entertainment since it would be com- two physical loudspeakers for XTC. [15] [23]
patible with existing stereo recordings. The previ-
ous general approach was to create virtual surround 2.2. Spatial Enhancement
sound speakers by simulating the effect of Head- The idea of spatial enhancement, or stereo im-
age widening, can be traced back to the time
Related Transfer Functions (HRTFs). Also the ideal
listening area of the virtual surround speakers is then when stereophonic audio first spread around the
very narrow in stereo loudspeaker reproduction, and world. Many early experiments were done in ar-
the effect is sensitive to an individual’s body shape. tificial stereophony and its optimization. See [11]
[5][15][27] This paper proposes a new approach that for review. Atal introduced a method to produce
addresses these problems. virtual widening of the loudspeakers for phantom
sources. [2] Virtual loudspeakers are located in be-
This present work aims to provide laptop entertain- tween the real loudspeakers. This was followed by
ers another option to deliver an improved overall im- the transaural system. [8] Based on the same idea,
mersive experience and relatively unrestricted range [16] proposed a stereo widening method that adds a
of motion without a complicated setup or required slightly attenuated, phase-shifted, and delayed ver-
additional equipment. The main algorithmic contri- sion signal to its opposite channel. This principle
bution leverages the combination of Ambiophonics is further elaborated by [1] and others to produce
principles [12] and dynamic Mid-Side techniques to a more natural stereo widening effect. As the mod-
project any stereo components playing in a laptop ern high-speed digital signal processing chips have
to a wider sound stage, on-the-fly, while keeping a gotten faster, it is possible to compute more com-
static center image. The present algorithm works plicated algorithms in real-time. For example, [26]
without any measured/synthesized HRTFs. A Mac proposed a real-time method that would map the
OS X menu bar application (Ambidio) is introduced. original stereo recording to virtual sound sources.
Finally, the performance of the present algorithm is When spatial enhancement is done during repro-
validated by comparing it with the competitive tech- duction, i.e. at the user end, it passively changes
nology, SRS iWOW, and traditional stereo in both the input stereo signal to increase the overall spatial
subjective and objective ways. perception without any special “guideline” from the
original audio producer. The method proposed in
2. BACKGROUND this paper belongs to this category.
2.1. Crosstalk Cancellation 2.3. Related Work
Crosstalk is an inherent problem in stereo loud- At the time of this writing, there are several com-
speaker playback. It occurs when a sound “falsely mercial products that have spatial enhancement fea-
reaches the ear on the opposite side from each tures implemented in different but also similar ap-
speaker.”[15] Unwanted spectral coloration is caused proaches. For example: products from Dolby Labs,
because of constructive and destructive interference QSound Labs, and Harman make use of general-
between the original signal and the crosstalk signal. ized HRTFs. [20][3][6] Products from DTS (for-
Moreover, conflicting spatial cues are created and merly SRS Lab), on the other hand, use frequency-
cause spatial distortion. As a result, localization and amplitude-dependent processing. Specifically,
fails and the stereo image collapses to the position the side signal is filtered by a curve in the shape
of the loudspeakers. The solution to this problem of average rear HRTFs. [17] Lastly, the BACCH
is a crosstalk cancellation (XTC) algorithm. XTC filter introduced by Princeton University is an op-
refers to adding a crosstalk canceling vector to the timal crosstalk cancellation filter that creates a flat
opposite speaker to acoustically cancel the crosstalk frequency response at the loudspeakers. Without
signal at a listener’s eardrum. With properly po- involving generalized HRTFs, it uses actual impulse
sitioned speakers, XTC also has the ability to en- responses measured from a dummy head or listener’s
hance traditional stereo recording by simulating vir- ear canals. [7]

AES 136th Convention, Berlin, Germany, 2014 April 26–29

Page 2 of 8
Wu et al. Ambidio: Sound Stage Width Extension for Internal Laptop Loudspeakers

Fig. 1: Modified RACE algorithm. Signal flow (solid line), user controlled parameters (dashed line)

All of the above mentioned technologies use either 3. IMPLEMENTATION

analyzed or measured HRTFs to enhance immersion. The name “Ambidio” suggests the underlying the-
As noted, HRTFs are very individualized, so none of ory—“Ambiophonics Audio.” Ambidio is a stereo-
the generalized HRTFs can compete with a listener’s in-stereo-out program mainly written in C++ and
own. Listening with someone else’s “ear” can cause developed on Mac OS X 10.8. The program can
spatial distortion resulting degraded sound localiza- be divided into four parts: First, a kernel is imple-
tion. Furthermore, since the HRTFs have fixed an- mented which serves as a virtual audio device that
gles, a listener needs to stay at the same fixed po- captures the system audio output. The captured
sition as the HRTFs to get the benefit of the XTC. output is then routed to a separate application in
If a listener moves away from the sweet spot, which the user space where the present spatial enhance-
accordingly changes the HRTFs necessary to cancel ment algorithm (Section 3.1) is applied. To keep
the crosstalk, the image collapses. [4] flexibility, the default output can be either laptop
2.4. Ambiophonics internal speakers or external speakers. We also pro-
Introduced by Ralph Glasgal in the 1980s, Ambio- vided an intuitive user interface. (Section 3.2)
phonics is an HRTF-free method taking advantage 3.1. Crosstalk Cancellation Filter Design
of psychoacoustic principles to project an existing As illustrated in Fig. 1, a spatial ratio is calcu-
stereo file up to a 180° stage by two closely spaced lated from the correlation between channels. The
(10-20°) frontal speakers and a XTC filter. Am- stereo inputs are first sent to a mixing block, which
biophonics is driven by a Recursive Ambiophonic balances the center image and the ambience sound
Crosstalk Eliminator (RACE) algorithm, which can- based on a calculated spatial ratio and selected
cels the crosstalk signal from the opposite speaker thresholds. By doing this, the side components are
recursively. See [12][13] for a detailed theory of Am- sure to be expanded efficiently by the XTC filter.
biophonics. The signals are then passed to the heart of the RACE
Techniques based on Ambiophonics have several ad- algorithm—invert, attenuate, and delay. Parame-
vantages. Foremost, there is no 30°speaker at each ters are optimized to match the physical configura-
side, so there is no possibility to make the sound tion of the hardware.
more like two point sources at the speakers. Also, After being processed by the RACE algorithm, the
the ideal listening area is wider and more robust depth and the room ambience of a stereo stream that
with respect to the listener’s head movement not were once buried under the collapsed image is recov-
only because the two loudspeakers are placed close ered. With such a feature, the audio content poten-
to each other [25], but also because there is no syn- tially appears to be farther in distance. Note that
thesized/measured HRTFs involved. Hence, it al- the use of artificial reverberation or even a small pan
lows listeners to listen with their own ears. Since from the center would make the center image drift
the separation angle of laptop internal speakers is to the side. Therefore, the RACE-enabled output is
always smaller than the traditional 60° listening tri- adjusted based on the spatial ratio and a selected
angle, the laptop configuration is well suited for the threshold. This keeps the center image strong and
application of the Ambiophonics techniques. stable enough for listeners, as it is an important fea-

AES 136th Convention, Berlin, Germany, 2014 April 26–29

Page 3 of 8
Wu et al. Ambidio: Sound Stage Width Extension for Internal Laptop Loudspeakers

ture to make the content understandable. Next, an adjust the height of the chair, and sit in any posture
EQ is applied to eliminate the audible coloration in or distance they were comfortable with.
high frequency bands created by using non- ideal fac-
4.1.2. Process Methods
tors in respect to the size of the head and the laptop.
The evaluation included three different audio pro-
Finally, a gain control block makes sure every signal
cessing methods: (a) a reference without any pro-
is within the proper amplitude range and is at the
cessing; (b) the competitive technology, SRS iWOW
same loudness as the original input signal.
(version 3.3); (c) the main output of this present
The present algorithm can also be implemented in work, Ambidio. SRS iWOW (chosen mainly because
the frequency domain to improve computational per- of its availability) was played using the default lap-
formance. Unlike having infinite recursive crosstalk top built-in speaker profile as provided by the man-
cancellation in time domain processing, it only re- ufacturer. Ambidio, on the other hand, used the de-
duces the crosstalk signals to -90 dB, yet still enough fault value as it came out of the spatial enhancement
to make the crosstalk inaudible. algorithm with no extra center image adjustment.
3.2. User Interface Design 4.1.3. Stimuli
Ambidio has a very simple and intuitive user inter- Four 20-second sequences were selected for the ex-
face. After launching the application, a small Am- periment. To eliminate the inherent stage difference
bidio icon appears on the menu bar. Clicking on the between sequences, all the sequences contained no-
icon can bring up a drop-down menu, where users ticeable hard-panned sounds. It was designed to in-
can quit the program or open the control panel. An clude different types of popular media entertainment
advanced control panel is also available that allows experiences, including game, movie and music. As
users to optimize the effect. a result, there were a total of 12 clips (3 methods
x 4 sequences), and 176 trials (44 subjects x 4 se-
4. SUBJECTIVE EVALUATION quences). All clips were prerecorded by ProTools 10
and normalized to the same B-weighted sound pres-
4.1. Subjective Evaluation Design sure level to prevent any bias created by level dif-
The subjective experiment was designed following ferences as suggested in [22]. The normalized clips
the paradigms of [22] and [10]. were stored as 44.1 kHz, 24 bit .wav files.
4.1.1. Participants and Apparatus
Forty-four graduate students and faculty members Name Content Resource Video
from New York University (17 females, 27 males, Game BioShock Infinite gameplay v
M±SE=27.0±5.0) voluntarily participated and were Movie Jurassic Park (1993) YouTube v
assigned to the same task. All participants had nor- Choir Banquet Fugue CD
mal hearing. Seventy three percent of the subjects Jazz Jazz music downloaded
were music majors or worked in a music-related field.
The experiment took place individually in the Spa- Table 1: Description of the Selected Sequences
tial Auditory Research Lab of the Music Technology
program at NYU. It was run on a MacBook Pro 15- 4.1.4. Procedure
inch. Participants were seated on a swivel chair. Custom software was written to run and guide the
They were allowed to move freely, turn their head, subjects through the entire listening test. (Fig. 3) It
provides interactive figures that help subjects to vi-
sualize a rating. The program randomly determines
the order of the four sequences. In each trial, three
clips with the same sequence were processed with
different methods in a random order. Subjects were
asked to judge the (a) stage width, (b) depth (per-
spective), and (c) presence (nearness) of each clip.
Fig. 2: User Interface of Ambidio Dummy variables were created to represent different
levels of perceived depth/presence.

AES 136th Convention, Berlin, Germany, 2014 April 26–29

Page 4 of 8
Wu et al. Ambidio: Sound Stage Width Extension for Internal Laptop Loudspeakers

Fig. 3: The custom software used in the experiment

In addition, subjects also had to choose one clip that 4.2.1. Perceived Width, Depth, and Presence
(d ) has the best sound quality; (e) is the most im-
Considering only the trials where the reference clip
mersive, and (f ) is their favorite. They were allowed
is correctly rated, the perceived stage width is
to listen to the clips more than once and go back
176±3° for Ambidio. (N=44) The stage width
and forth between trials until they felt satisfied with
boost was analyzed to eliminate the impact of
their answer.
the presentation order, the sequence types, and
the differences within subjects. The average stage
4.1.5. Data Analysis width boost of Ambidio is significantly larger than
All the data were analyzed at an alpha level of .05. SRS iWOW. (F (3,172)=5.15, p<.001) In addi-
Levene’s test was used to examine the variance’s ho- tion, Ambidio works equally well on four sequences
mogeneity whitin individual group. Either N-way (F (3,172)=1.49, p=.22), whereas significant differ-
ANOVA or Brown-Forsythe’s test was used depend- ences were observed between sequences when the
ing on the homogeneity. If the F value is larger than method is SRS iWOW. (F (3,172)=5.15, p<.001)
alpha, a post-hoc test would be performed by either
Fisher’s LSD or Scheffe’s method depending on the On the depth rating, 44% subjects rated the per-
sample sizes for each group. ceived depth as “far” (score≥2) when the process
method is Ambidio, compared to 23% and 18% for
4.2. Subjective Evaluation Result SRS iWOW and the reference clip, respectively. .
No significant effect was found for subjects’ back- Ambidio provided deeper depth than the other two
ground or the presence of the accompanying video methods. (p<.05) The rated depth did not signif-
for all rated features. There was no significant dif- icantly differ between SRS iWOW and the hidden
ference between the sequences unless otherwise spec- reference. (p=.06) For the presence rating, 68% par-
ified. The results are reported as M±SE. ticipants rated the presence as “close” (score≥2) in
clips processed by Ambidio, compared to 37% and

Score Depth Presence Ambidio SRS iWOW reference

0 right at the speakers Average Absolute Width 176±3° 74±6° 59±0.7°

1 a bit behind the screen a bit in front of the screen Average Stage Boost 111±4° 19±4° -
2 far, but can be farther close, but can be closer Average Depth 1.32±0.07 0.86±0.06 0.75±0.06
3 as far as it can be right at my ears Average Presence 1.80±0.07 1.23±0.07 0.82±0.06

Table 2: Dummy Variables for Depth and Presence Table 3: Perceived Rating Results (N=44, n=176)

AES 136th Convention, Berlin, Germany, 2014 April 26–29

Page 5 of 8
Wu et al. Ambidio: Sound Stage Width Extension for Internal Laptop Loudspeakers

21% in clips processed by SRS iWOW and the ref- Quality A A A A A

Immersion B B B B A
erence clip, respectively. The simple main effect Preference A B C tie A
test revealed that the perceived presence provided Count
Quality
84 (48%)
Immersion Either
12 (7%)
- Ambidio
80 (45%)
SRS iWOW reference
by Ambidio was significantly closer than the other 30 (36%) 51 (61%) 3 (4%) - 79 (99%) 1 (1%) 0 (0%)

two methods. (p<.001) Table 4 shows the descrip-

tive results of each rated feature. Table 6: All Possible Combinations for Multiple
Choices (N=44, n=176)
Ambidio SRS iWOW reference tie
Best Width 93% 2% 0% 5% 5.1. Objective Measurement Design
Best Depth 41% 8% 10% 41% An objective experiment was conducted to examine
Best Presence 52% 12% 3% 33% the frequency response of Ambidio. The experiment
took place in the same room as described above, and
Table 4: Best Rating within Trials (N=44, n=176) used the same apparatus as in subjective evaluation.
A Neumann KU100 dummy head was used. It was
set at the same height as the author’s head facing
4.2.2. Sound Quality, Immersion, and Preference
down to mimic a listener watching the screen. Test
Table 5 summarizes the result of the multiple-choice
signals were generated in 24-bit, 44.1 kHz: (a) iden-
selections. The preference selection data were ana-
tical pink noise in left/right channels (coherent sig-
lyzed to examine their relationship to the best sound
nal for center image); (b) same as (a), but with the
quality selection and the most immersive selection.
right channel at -6 dB (intermediate); (c) same as
As shown in Table 6, there were 84 trials where a
(a), but with the right channel 180° phase inverted.
subject was asked to choose between the sound qual-
[22] A 20-second white noise signal was also gener-
ity and the immersion. 61% preferred the latter one.
ated in the same format.
There were 80 trials in which a single clip was voted
to have the best sound quality, to be the most im- Two impulse responses corresponding to each ear
mersive, and to be the preferred clip. Among 80 were recorded separately. (as shown in Fig. 4 top)
trials, 79 of them were Ambidio. The recorded data was computed by 214 point FFT
with Blackman-Harris window for harmonic isola-
Ambidio SRS iWOW reference tie tion. The resulting spectrum was normalized and
Best Sound Quality 46% 30% 13% 11% deconvolved using the corresponding impulse re-
Best Immersion 94% 3% 1% 2% sponse. A 1/24th -octave filter was applied. If the
Preference 73% 15% 5% 7% algorithm introduced no coloration and eliminated
all the crosstalk, the deconvolved result should be
Table 5: Best Rating within Trials (N=44, n=176) identical to the original sound file.
5.2. Objective Measurement Result
5. OBJECTIVE MEASUREMENT Fig. 5 shows the frequency responses of each test
clip. Frequencies outside a window of 100-10kHz

Fig. 4: Illustration of the Signal Chain in the Objective Experiment

AES 136th Convention, Berlin, Germany, 2014 April 26–29

Page 6 of 8
Wu et al. Ambidio: Sound Stage Width Extension for Internal Laptop Loudspeakers

Fig. 5: The measured response. red: digital file; blue: reference; green: SRS iWOW; orange: Ambidio. The
area covered in gray is too low/high for a laptop to reproduced, so they are not to be taken as shown.

may have some distortions, due to the laptop’s in- One of the limitations of Ambidio is that it does not
ability to reproduce these frequencies accurately. As support headphones since the underlying theories do
mentioned before, the ultimate goal is to fit the red not apply. This can still be done by simulating vir-
line in Fig. 5, i.e. the original sound file. The error tual speakers for headphones. Moreover, when there
of the stereo reference was significant in all of the is no interaural differences available (both channels
three test clips. It is also noteworthy that the three identical), no realistic stage width can be auralized.
left curves for Ambidio were similar.
With the increasing importance of media entertain-
6. DISCUSSION AND CONCLUSIONS ment using consumer-grade hardware, passive spa-
The main results of the present paper show that tial enhancement techniques have the potential to
Ambidio had an average stage width boost of 111°. expand the sound stage during playback, thus pro-
(N=44) ) This could be the reason that it is neither viding a more immersive listening experience for
based solely on conventional widening methods, nor gaming, movies, or pure music listening purposes.
HRTFs. In addition, Ambidio not only performed This paper introduces a method, Ambidio, to extend
better in depth and presence rating (Table 3), but the width of the sound stage for built-in laptop loud-
also picked as the favorite clips among 73% of 176 tri- speakers. Ambidio relies on the traditional stereo
als. Also revealing from the selection result, widen- inputs and does not need additional preprocessing
ing the stage width could be the easiest way to create to extend the perceived stage width, immersiveness
a feeing of immersion as nearly all (95%) of the clips and realism. A subjective experiment was conducted
selected as the most immersive were providing the in which Ambidio was compared with a commercial
widest soundstage width. On the other hand, the ob- spatial enhancement program—SRS iWOW by 44
jective results show that spatial distortion appears subjects in an acoustic controlled environment us-
to be reduced when crosstalk signals are minimized, ing laptop internal speakers. Results from subjec-
leading to more successful spatial enhancement. tive tests indicate that the method presented here
shows promise in significantly extending the sound
Immersion is an important feature for laptop enter- stage width, increasing the sense of immersion and
tainment, as underlined by results in the experiment minimizing spectral coloration.
described above where most subjects chose immer-
sion rather than sound quality as the most impor-
tant feature. It is noteworthy that one subject men- 7. ACKNOWLEDGEMENTS
tioned: “I already got used to listening to that kind The author would like to thank Pei-Lun Hsieh, Shao-
of sound (of internal speakers).” It suggests that lis- Ting Sun, and Cheng-I Wang for all the helpful con-
teners may need time to adapt themselves to listen versations. Special thanks to Hao Li for his advice
to sound that isn’t purely coming from the front. and comments on the manuscript.

AES 136th Convention, Berlin, Germany, 2014 April 26–29

Page 7 of 8
Wu et al. Ambidio: Sound Stage Width Extension for Internal Laptop Loudspeakers

8. REFERENCES [15] G. S. Kendall. A 3-d sound primer: directional

[1] R. M. Aarts. Phantom sources applied to hearing and stereo reproduction. Computer mu-
stereo-base widening. Journal of the Audio En- sic journal, 19(4):23–46, 1995.
gineering Society, 48(3):181–189, 2000. [16] W. Kitzen and P. Boers. Applications of a
[2] B. S. Atal. Apparent sound source translator, digital audio-signal processor in tv sets. In
1966. US Patent 3,236,949. Acoustics, Speech, and Signal Processing, IEEE
International Conference on ICASSP’84., vol-
[3] J. Bauck and D. H. Cooper. Generalized ume 9, pages 527–530. IEEE, 1984.
transaural stereo and applications. Journal of
the Audio Engineering Society, 44(9):683–705, [17] A. I. Klayman. Stereo enhancement system,
1996. 1988. US Patent 4,748,669.

[4] D. R. Begault. 3D-sound for virtual reality [18] A. Kraemer. Two speakers are better than 5.1
and multimedia. Academic Press Professional, [surround sound]. Spectrum, IEEE, 38(5):70–
Boston, 1994. 74, 2001.

[5] J. Blauert. Spatial hearing: the psychophysics [19] Logitech and Wakefield Research. Survey finds
of human sound localization. MIT press, 1997. our close relationship with our laptops isn’t
without issues, 2010.
[6] C. P. Brown. Surround sound virtualizer and
method with dynamic range compression, 2011. [20] D. D. Lowe and J. W. Lees. Sound imaging
US Patent 2011/0,243,338. process, 1991. US Patent 5,046,097.
[7] E. Choueiri. Spectrally uncolored optimal [21] R. C. Maher, E. Lindemann, and J. Barish. Old
croostalk cancellation for audio through loud- and new techniques for artificial stereophonic
speakers, 2013. US Patent 2013/0,163,766. image enhancement. In Audio Engineering So-
ciety Convention 101, 1996.
[8] D. H. Cooper and J. L. Bauck. Prospects for
transaural recording. Journal of the Audio En- [22] S. Olive. Evaluation of five commercial stereo
gineering Society, 37(1/2):3–19, 1989. enhancement 3d audio software plug-ins. In Au-
dio Engineering Society Convention 110, 2001.
[9] Dolby Laboratories. Dolby survey finds enter-
tainment features become must-have on pcs for [23] F. Rumsey. Spatial audio. Focal Press, 2001.
college students, 2010.
[24] M. R. Schroeder. Listening with two ears. Mu-
[10] Dolby Laboratories. Pc postprocessing tech- sic perception, pages 255–280, 1993.
nologies: A competitive analysis, 2011.
[25] T. Takeuchi, P. A. Nelson, O. Kirkeby, and
[11] J. Eargle, editor. Stereophonic Techniques: An H. Hamada. Robustness of the performance of
anthology of reprinted articles on stereophonic the” stereo dipole” to misalignment of head po-
techniques. Audio Engineering Society, 1986. sition. In Audio Engineering Society Conven-
[12] R. Glasgal. 360 ◦ localization via 4. x race pro- tion 102, 1997.
cessing. In Audio Engineering Society Conven- [26] C. Tsakostas, A. Floros, and Y. Deliyiannis.
tion 123, 2007. Binaural rendering for enhanced 3d audio per-
[13] R. Glasgal. Ambiophonics: Beyond Surround ception. In Proceedings of the AudioMostly 2007
Sound to Virtual Sonic Reality. Ambiophonics 2nd Conference on Interaction with Sound,
Institute, New Jersey, 2nd edition, 2009. pages 27–29, 2007.

[14] Google. The new multi-screen world: Un- [27] W. A. Yost. Fundamentals of Hearing: An In-
derstanding cross-platform consumer behavior, troduction. Academic Press, 4th edition, 2007.
2012.

AES 136th Convention, Berlin, Germany, 2014 April 26–29

Page 8 of 8