References

The Sonic Arts Research Centre. (2014). Retrieved January 2014, from The Sonic Arts Research Centre, QUB: http://www.sarc.qub.ac.uk/

The Ulster String Quartet. (2014). Retrieved September 2013, from The Ulster String Quartet: http://ulsterquartet.com

Bech, S., & Zacharov, N. (2006). Perceptual Audio Evaluation – Theory, Method and Application. John Wiley & Sons, Ltd.

Berge, S., & Barett, N. (2010a). A New Method for B-format to Binaural Transcoding. 40th AES International Conference. Tokyo, Japan: Audio Engineering Society.

Berge, S., & Barret, N. (2010b). High Angular Resolution Planewave Expansion. 2nd International Symposium on Ambisonics and Spherical Acoustics, (pp. 6-7). Paris, France.

Blumlein, A. D. (1931). Patent No. British Patent Specification 394,325.

Bower, K. (2014). Worked Examples. Retrieved February 2014, from Keith Bower: http://www.keithbower.com/Worked_Examples.htm

Choisel, S., & Wickelmaier, F. (2007). Evaluation of Multichannel Reproduced Sound: Scaling Auditory Attributes Underlying Listener Preference. The Journal of the Acoustical Society of America, 388-400.

Conover, W. J. (1999). Practical Nonparametric Statistics (3rd ed.). John Wiley & Sons, Inc.

Cox, T. (2013). Objective Metrics. Retrieved from Salford University – Acoustics, Audio and Video: http://www.acoustics.salford.ac.uk/acoustics_info/sound_quality/?content=objective

Cox, T. (2013). Subjective Methods. Retrieved September 25, 2013, from Salford University – Acoustics, Audio and Video: http://www.acoustics.salford.ac.uk/res/cox/sound_quality/index.php?content=subjective

Craven, P. G., & Gerzon, M. A. (1977). Patent No. 4,042,779. United States of America.

Cycling 74. (2014). Max. Retrieved January 2014, from Cycling 74: http://cycling74.com/products/max/

DV247. (2014). AKG C414 XLII. Retrieved April 5, 2014, from Microphones: http://www.dv247.com/microphones/akg-c-414-xl-ii–72454

Eargle, J. (2005). The Microphone Book. Focal Press.

e-Discovery Team. (2014). Uploads. Retrieved March 2014, from e-Discovery Team: http://ralphlosey.files.wordpress.com/2012/05/95percent_confidence_normaldist1-96.png

Elder Laboratory. (2014). The Binomial Test. (T. C. York University, Producer) Retrieved November 2013, from Elder Laboratory: Human & Computer Vision: http://www.elderlab.yorku.ca/~aaron/Stats2022/BinomialTest.htm

European Broadcasting Union. (2011, August). Recommendation R128 – Loudness normalisation and permitted maximul level of audio signals. Retrieved September 2013, from European Broadcasting Union: https://tech.ebu.ch/docs/r/r128.pdf

European Commission. (2014). Scientific Committees. Retrieved January 2014, from Scientific Committee on Emerging and Newly Identified Health Risks: http://ec.europa.eu/health/scientific_committees/emerging/index_en.htm

Fazenda, B. (2012). Lecture Material. M.Sc. Digital Media: Audio Production – Audio Post Production. United Kingdom: University of Salford.

Focusrite. (2014). Saffire Pro 24 Technical Specifications. Retrieved March 2014, from Focusrite: http://us.focusrite.com/firewire-audio-interfaces/saffire-pro-24/specifications

Fukada, A. (2001, June). A Challenge in Multichannel Music Recording. Audio Engineering Society, Paper Number 1881.

Fukada, Tsujimoto, & Akita. (1997, September). Microphone Techniques for Ambient Sound on a Music Recording. Audio Engineering Society, Paper Number 4540.

Genelec. (2014). 1032a Product page. Retrieved October 2013, from Genelec: http://www.genelec.com/products/previous-models/1032a/

Gerzon, M. A. (1975). The Design of Precisely Coincident Microphone Arrays for Stereo and Surround Sound. Journal of the Audio Engineering Society, Prepint L-20.

Goldman, M. (2014). Why is mulitiple testing a problem? Retrieved November 2013, from Berkeley University of California: http://www.stat.berkeley.edu/~mgoldman/Section0402.pdf

Google. (2014). Google Maps. Retrieved October 2013, from Google.ie: https://maps.google.ie/maps?q=queens+university+belfast&hl=en&sll=53.3834,-8.21775&sspn=9.024249,25.378418&t=h&z=20&iwloc=A

Harpex. (2011). Harpex B Manual. Retrieved September 2013, from Harpex: http://harpex.net/manual.pdf

Harpex. (2014). About. Retrieved February 2014, from Harpex: http://harpex.net/about.html

Harpex. (2014). Index. Retrieved November 2013, from Harpex: http://harpex.net/

Heck, P., & Rieseberg, C. (2001). Comparison of the Four Different Microphone Arrays. Duesseldorf Universirty, Duesseldorf.

HHB Communications Ltd. (2014). Soundfield MKV. Retrieved April 5, 2014, from HHB Communications Ltd: http://www.hhb.co.uk/product_detail.php?id=1753

Holman, T. (2010). Sound for Film and Television. Taylor & Francis.

Houghton, M., & Sound on Sound. (2011, August). Sound Advice. Retrieved January 2014, from Sound on Sound: http://www.soundonsound.com/sos/aug11/articles/qanda-0811-3.htm

Howard, D. M., & Angus, J. A. (2009). Acoustics and Psychoacoustics (4th ed.). Oxford: Focal Press.

ITU. (1997, 10). BS.1116-1. Retrieved January 2014, from International Telecommunication Union: https://www.itu.int/dms_pubrec/itu-r/rec/bs/R-REC-BS.1116-1-199710-I!!PDF-E.pdf

ITU. (2003, 01). BS.1534-1. Retrieved January 2014, from International Telecomunications Union: http://www.itu.int/dms_pubrec/itu-r/rec/bs/R-REC-BS.1534-1-200301-I!!PDF-E.pdf

ITU. (2012, August). BS.775-3. Retrieved September 2013, from International Telecommunication Union: http://www.itu.int/dms_pubrec/itu-r/rec/bs/R-REC-BS.775-3-201208-I!!PDF-E.pdf

Kassier, R., Hyun-Kook, L., Brookes, T., & Rumsey, F. (2005). An Informal Comparison Between Surround-Sound Microphone Techniques. The Journal of the Audio Engineering Society, Paper Number 6429.

Katz, B. (2007). Mastering Audio: The Art and the Science (2nd ed.). Focal Press.

Kelly, R. (2014, January). Listening Test Preparation. Retrieved January 2014, from Rob Kelly Sound: https://robkellysound.wordpress.com/2014/02/01/listening-session-preparation/

Kornacki, A., Odya, P., Kostek, B., & Czyzewski, A. (2001, May). Problems Related to Surround Sound Production. Journal of the Audio Engineering Society, Paper Number 5374.

Kuster, M. (2008). Reliability of Estimating the Room Volume from a Single Room. The Journal of the Acoustical Society of America 124(2), 982-993.

Microsoft. (2014). NORM.S.INV function. Retrieved March 2014, from Microsoft Office: http://office.microsoft.com/en-gb/excel-help/norm-s-inv-function-HP010335692.aspx

Minitab. (2014). Minitab. Retrieved March 2014, from Minitab: http://www.minitab.com/en-us/products/minitab/

Moylan, W. (2007). Understanding and Crafting the Mix – The Art of Recording. Focal Press.

Paquier, M., & Koehl, V. (2011). Subjective Assessment of Microphone Arrays for Spatial Audio Recording. Forum Acusticum 2011, (pp. 2737-2742). Aalborg.

Pekonen, J. (2008). Microphone Techniques for Spatial Sound. Retrieved October 2013, from Academia.edu: www.academia.edu/download/30852150/Pekonen2008ACOUSTICS.pdf

Pohlmann, K. C., & Everest, E. A. (2001). Master Handbook of Acoustics. McGraw Hill.

Preston, C. (2003). Adaptations of Stereo Microphone Techniques for Surround Sound. AES 24th International Conference: Multichannel Audio, The New Reality, (Paper 2). Banff.

Preston, C. (2005, October). Alternative Approaches for Recording Surround Sound. AES Convention 119, (Paper Number 6623). New York.

Queen’s University Belfast. (2014). School of Music. Retrieved January 2014, from School of Creative Arts: http://www.music.qub.ac.uk/

Recording Hacks. (2014). Neumann M50. Retrieved March 2014, from Recording Hacks: http://recordinghacks.com/microphones/Neumann/M-50

Robjohns, H., & Sound on Sound. (2005, December). Q&A. Retrieved January 2014, from Sound on Sound: http://www.soundonsound.com/sos/dec05/articles/qa1205_4.htm

Rumsey, F., ZielinskiI, S., & Bech, S. (2008). On Some Biases Encountered in Modern Audio Quality Listening Test. The Journal of the Audio Engineering Society, 56(6), 56, 427 – 451.

Salford University. (2014). Recording Studios. Retrieved March 2014, from Acoustics, Audio and Video: http://www.acoustics.salford.ac.uk/facilities/?content=studios

Schellstede, M., & Faller, C. (2007). Recording of Acoustical Concerts Using a Soundfield Microphone. AES Convention 112, (Paper Number 6996). Vienna.

Sengspiel, E. (2014). Calculation of the Reverberation Time. Retrieved November 2013, from Sengspiel Audio: http://www.sengpielaudio.com/calculator-RT60.htm

Soundfield Ltd. (2013). Soundfield MKV. Retrieved August 2013, from Soundfield Ltd.: http://www.soundfield.com/products/mkv.php

Soundfield Ltd. (2014). Benefits of a Soundfield System. Retrieved September 2013, from Soundfield: http://www.soundfield.com/soundfield/soundfield.php

Soundfield Ltd. (2014). Software Plugins. Retrieved September 2013, from Soundfield Ltd.: http://www.soundfield.com/products/plugins.php

Soundfield Ltd. (2014). The Company – background and history. Retrieved October 2013, from Soundfield: http://www.soundfield.com/company/company.php

Sprent, P. (1989). Applied Nonparametric Statistical Methods. Chapman and Hall.

Sungyoung, K., Franscisco, M. d., Walker, K., Marui, A., & Martens, W. L. (2006). An Examination of the Influence of Musical Selection on Listener Preferences for Multichannel Microphone Technique. Audio Engineering Society Conference: 28th International Conference: The Future of Audio Technology–Surround and Beyond (Paper Number 9-2). Pitea: Audio Engineering Society.

Tischmeyer, F. (2012, August 14). Steinberg SLM 128-Loudness Meter presented by Friedemann Tischmeyer . Retrieved October 26, 2013, from YouTube: https://www.youtube.com/watch?v=w0yrb61f8_k

University, West Chester. (2014). Retrieved March 2014, from WCU Course Server: http://courses.wcupa.edu/rbove/Berenson/10th%20ed%20CD-ROM%20topics/section6_5.pdf

Williams, M., & Le Du, G. (1999, September). Microphone Array Analysis for Multichannel Sound Recording. Audio Engineering Soceity Convention 107, (Paper Number 4997).

Williams, M., & Le Du, G. (2000, February). Multichannel Microphone Array Design. Audio Engineering Society Convention 108, (Paper Number 5157). Paris.

Wuttke, J. (2005, October). Surround Recording of Music: Problems and Solutions. Audio Engineering Society Convention 119 (Paper Number 6556). New York: Audio Engineering Society.

Appendix A.       Max Patch

Figure A.1 – Test Interface

The test interface, shown in Figure A.1, was built in a Max 6.1 (Cycling 74, 2014). A number of elements had to be designed in order to meet the aims of the test which can be summarised as to:

1. Provide a test interface which will allow for pairwise comparison of two surround sound recording extracts,
2. Provide a means to randomise playback of test material,
3. Provide looped playback of test material,
4. Provide real time switching of test material,
5. Allow for participants to submit their answers to test questions,

Once a system which meets the above criteria is established, it can then be copied and pasted as required. For example, by building and troubleshooting one test section, the final result can be copied and patched in with ease. This Appendix outlines the design of the major components the patch.

A.1         Randomisation

To meet the randomisation section of the ITU-R BS.1116-1 recommendation, a way of randomising the playback of test stimuli had to be incorporated into the test interface.

Figure A.2 shows one of the twelve components of the randomising system of the patch. Each of the audio extracts for each pairwise comparison was fed into one of these randomising sections. The objects ‘open sf1.aif’ and ‘open t1.aif’ correspond to the Soundfield ITU and MMA recordings of the first section of musical piece 1. This file information is then routed to the sound file players via a randomising system.

Figure A.2 – Randomisation

By entering a randomisation code, where each number is applicable to each randomising section, the destination of each array pair could be randomised. For instance, if Pair A is played back first by default and if the first number of the randomising code was eight then Pair A would be the eighth pairwise comparison presented in the test. Table A.1 shows how a full randomising code changes the default playback order of the test with arrows highlighting two of the playback order changes.

Table A.1 – Randomising Operation

This system allows the test supervisor to input a code from a pool of randomising codes before participants begin the test. The randomising code is presented with the participant’s answers so they can be integrated into the master spread sheet where the reverse of the randomisation process is applied to allow for the analysis of the results. To avoid using the default playback order by accident, the patch can only be initialised after the randomisation code is confirmed as being entered correctly.

With reference to Figure A.2, the twelve number buttons which are routed inputs one to twelve inclusive and inputs thirteen to twenty-four inclusive of a sub-patch denoted by the name ‘p 14’. The sub-patch, shown in Figure A.3, routes the signals from its inputs into two gate objects. There is one gate object per recording extract. Sound file information from the array pair enter the sub-patch through inputs twenty-five and twenty-six.

Figure A.3 – Gates

With the randomising number given, the number acts on both the gates and sends the sound file information for the recording extracts through the appropriate gate output. For example, if number one is pressed, the signal of input 25 and 26 will be sent out the second output of gates one and two respectively. The addition of the one which can be seen at the top of Figure A.3 is to eliminate errors further down the playback chain. The gate outputs are routed to the sub patch outputs and routes signals to the ‘send’ sub patch shown in Figure A.4

The outputs of the gates are then routed to the appropriate send objects. The send object allows a signal to be sent around a patch without the use of patch cords. If the number one is pressed, the gates will send the sound file information for the array pair to ‘send 1a’ and ‘send 1b’ respectively.

Figure A.4 – Sends

A.2         Playback

Figure A.5 shows how the sound file information is processed. The objects called ‘receive 1a’ and ‘receive 1b’ receive the signals sent by the objects ‘send 1a’ and ‘send 1b’ respectively. The receive objects are then routed to two separate objects called ‘sfplay~’ which is a sound file player. The signals which have been sent as outlined in Section A.1 correspond to the file locations of the tests recording extracts. The ‘sfplay~’ object will then use this information to open the appropriate files for playback. The first argument for the ‘sfplay~’ object defines how many channels are required for playback. A 5.1 surround sound channel requires 6 channels, hence ‘sfplay~ 6’.

Figure A.5 – Receive

Figure A.6 outlines the playback stage. The ‘Play’ and ‘Stop’ buttons are visible to the participant as controls of the audio playback. By pressing play, the playback loop feature of the ‘sfplay~’ object is activated.

Figure A.6 – Playback

The sets of outputs from the two ‘sfplay~’ objects are routed to inputs two and three of the six ‘selector~’ objects respectively. The ‘Recording 1’ and ‘Recording 2’ objects are visible to the test participants. These are seen by participants as sound selection controls which allow for the real time switching of audio file playback. The output of these buttons is routed into a ‘1’ and ‘2’ message objects respectively.

The message boxes are in turn fed into all of the first inputs of the ‘selector~’ objects. If the selector objects receive a ‘1’, they will send the signal from the second input through to its output. If they receive the number ‘2’, they will send the signal from their third input to their output. This group action switches between the output of either ‘sfplay~’ object.

The ‘delay’ sub patch sends a ramped signal between one and zero out to the six multiplication objects which are placed before the audio output, called the ‘dac~’ object. When the participant selects a recording, the outputs are muted by the delay sub patch. The sub patch then sends out the switching signal before unmuting the audio outputs.

Figure A.7 – Playback Display

Figure A.7 shows a subsection of Figure A.6. The outputs of the ‘Recording 1’ and ‘Recording 2’ objects send a one or zero depending on their on/off state. They are routed to ‘if’ statements. The ‘if’ statements are then routed to a display object. The display object will display a separate message for each of the input signals ‘0’ and ‘1’. The display object is visible to the participant and provides them with the information of what recording extract they are listening to.

A.3         Answer Submission and Collection

Figure A.8 shows an example set of answers. The objects ‘Recording 1’ and ‘Recording 2’ are visible to the participants in their respective sections of the visual test interface. The two sets of answers displayed in Figure A.8 correspond to the preference and spaciousness questions of a test section. If one answer is given by a participant, the output of each answer section will be one. If no or two answers are given, the output will be zero or ten respectively.

Figure A.8 – Answer Collection

In the case of Figure A.8, the desired output of one and undesired output of ten have been sent to the answer verification section of the patch which is shown in Figure A.9. From the five test questions, the section of the patch is reacting to whether it receives a number five or not. In this example, question one was answered acceptably where question two was not with the rest of the questions unfilled.

Figure A.9 – Answer Verification 1

The ‘expr’ object adds up the output of each answer section and sends it to an if statement. As the output does not equal five, a warning message is sent to the participant instructing them to amend their answers before continuing. The desired situation of all questions being answered correctly is shown in Figure A.10 where all five questions have been properly filled out. The result of five from the ‘expr’ calculation gives an instruction to the participant to move to the next test section or contact the test supervisor where appropriate.

Figure A.10 – Answer Verification 2

Appendix B.       Result Data

Appendix section B.1 details the two formulae used to calculate the Z-score and p-value of the significance tests. The remainder of the Appendix provides the results of the pairwise comparisons which were graphically presented in Section 6.

B.1         Formulae

B.1.1      One Proportion Binomial

As the data from the pairwise comparison in this project is categorical, the one proportion binomial test can be used to calculate significance between the proportions of the comparisons sample (Bower, 2014)(Elder Laboratory, 2014).

To use an example, a set forty participants were presented with a pairwise comparison between options A and B and asked which one they preferred. The null hypothesis states that the options are equally preferred, thus H₀ equals 0.5. If thirty selected option A, then the number of ‘events’ would be thirty which equals x.

Equation B.1 – One Proportion Binomial Components and Example and Result

As the null hypothesis states that there is equal probability of A and B being preferred by participants, the two tailed test allows for the determination of the preference between A or B. This results in a critical region for a 95% significance level from -1.96 to 1.96. The binomial test revealed that there is a significant preference for option A at a 95% confidence level, z = 3.16, p < 0.05.

B.1.2      Bonferroni Correction

The hypothesis testing formulae used were integrated into the master Excel spread sheet which contained the test result data. The statistical software Minitab was used to confirm the correct operation of the spread sheet calculations (Minitab, 2014). Due to the number of comparisons being made as described in Section 4.1, a Bonferroni correction of three was applied (Goldman, 2014).

Equation B.2 – Bonferroni Correction 1

Equation B.2 outlines the Bonferroni correction where o is the original confidence level and b is the Bonferroni correction value. This calculation produces an adjusted significance level to take into account the number of comparisons being made per set of test stimuli. The significance levels are used to assess the results of certain hypothesis testing.

For example, if the result is below the significance level of 0.05 then the result is significant to a 95% confidence level. Similarly, if the result is below 0.01 then the result is significant to a 99% confidence level.

The binomial test does not produce p-values; however, it does operate with respect to the critical region described in Section 5.2 as the results of the test will either be inside or outside the region. The NORM.S.INV function of Excel, which calculates the critical region, requires a significance level to operate. The result of the Bonferroni correction can then be implemented into the spread sheet where the NORM.S.INV adjusts the critical regions which in turn adjust the results of the binomial test.

With a Bonferroni correction of three applied, Equation B.3 and Equation B.4 show the adjusted 95% and 99% significance levels respectively. These values have been used in the calculation of results of the project listening tests which are presented in Appendix Sections B.2, B.3 and B.4.

Equation B.3 – Bonferroni Correction 2

Equation B.4 – Bonferroni Correction 3

The original significance levels for 95% and 99% are 0.05 and 0.01 respectively which have now been adjusted to 0.017 and 0.003 respectively. These values are then used to calculate the critical regions for use in the binomial tests. For example, the original critical region for a 95% confidence level is -1.96 to 1.96. The Bonferroni adjusted critical region for the 95% confidence level is -2.39 to 2.39.

B.1.3      Chi-Square

To find the exact probability of a result in a pairwise comparison, known as the p-value, the Chi-Square test was used. In the example presented in Appendix B.1.1, the expected preference or probability that a population will select option A is equal. With a sample size of 40, the means the expected result for each category is 20. The observed result for category A is 30 and 10 for category B. For the result to meet the 95% confidence level, the result of the chi-square test must be less than the significance level of 0.05. The Chi-Square test was used, in addition to the Minitab software, to confirm the results of the binomial test.

Equation B.5 – Chi-square Example and Result

B.2         Overall Preference Results

B.2.1      MMA vs. SFI

65 out of 80 votes preferred MMA.  A binomial test revealed that there is a significant preference for MMA at a 99% confidence level, z = 5.59, p < 0.003.

B.2.2      MMA vs. SFA

50 out of 80 votes preferred MMA.  A binomial test revealed that there is not a significant preference for either array at a 95% confidence level, z = 2.24, p > 0.017.

B.2.3      SFI vs. SFA

29 out of 80 votes preferred SFI.  A binomial test revealed that there is a significant preference for SFA at a 95% confidence level, z = -2.46, p < 0.017

B.3         Preference Results by Piece

B.3.1      MMA vs. SFI

In piece 1, 34 out of 40 votes preferred MMA.  A binomial test revealed that there is a significant preference for MMA at a 99% confidence level, z = 4.43, p < 0.003.

In piece 2, 31 out of 40 votes preferred MMA.  A binomial test revealed that there is a significant preference for MMA at a 99% confidence level, z = 3.48, p < 0.003.

B.3.2      MMA vs. SFA

In piece 1, 22 out of 40 votes preferred MMA.  A binomial test revealed that there is not a significant preference for either array at a 95% confidence level, z = 0.63, p > 0.017.

In piece 2, 28 out of 40 votes preferred MMA.  A binomial test revealed that there is a significant preference for MMA at a 95% confidence level, z = 2.53, p < 0.017.

B.3.3      SFI vs. SFA

In piece 1, 14 out of 40 participants preferred SFI.  A binomial test revealed that there is not a significant preference for either array at a 95% confidence level, z = -1.90, p > 0.017.

In piece 2, 15 out of 40 participants preferred SFI.  A binomial test revealed that there is not a significant preference for either array at a 95% confidence level, z = -1.58, p > 0.017.

B.4         Attributes

B.4.1      MMA vs. SFI Piece 1 and Piece 2

B.4.1.1         Spaciousness

In piece 1, 28 out of 40 participants chose MMA.  A binomial test revealed that there is a significant preference for MMA at a 95% confidence level, z = 2.53, p < 0.017.

In piece 2, 26 out of 40 participants chose MMA.  A binomial test revealed that there is not a significant preference for either array at a 95% confidence level, z = 1.90, p > 0.017.

B.4.1.2         Envelopment

In piece 1, 31 out of 40 participants chose MMA.  A binomial test revealed that there is a significant preference for MMA at a 99% confidence level, z = 3.48, p < 0.003.

In piece 2, 28 out of 40 participants chose MMA.  A binomial test revealed that there is a significant preference for MMA at a 95% confidence level, z = 2.53, p < 0.017.

B.4.1.3         Clarity

In piece 1, 12 out of 40 participants chose MMA.  A binomial test revealed that there is a significant preference for SFI at a 95% confidence level, z = -2.53, p < 0.017.

In piece 2, 15 out of 40 participants chose MMA.  A binomial test revealed that there is not a significant preference for MMA at a 95% confidence level, z = -1.58, p > 0.017.

B.4.1.4         Naturalness

In piece 1, 23 out of 40 participants chose MMA.  A binomial test revealed that there is not a significant preference for either array at a 95 confidence level, z = 0.95, p > 0.017.

In piece 2, 21 out of 40 participants chose MMA.  A binomial test revealed that there is not a significant preference for either array at a 95 confidence level, z = 0.32, p > 0.017.

B.4.2      MMA vs. SFA Piece 1 and 2

B.4.2.1         Spaciousness

In piece 1, 22 out of 40 participants chose MMA.  A binomial test revealed that there is not a significant result for either array at a 95% confidence level, z = 0.63, p > 0.017.

In piece 2, 20 out of 40 participants chose MMA.  A binomial test revealed that there is not a significant result for either array at a 95% confidence level, z = 0.00, p > 0.017.

B.4.2.2         Envelopment

In piece 1, 27 out of 40 participants chose MMA.  A binomial test revealed that there is not a significant result for either array at a 95% confidence level, z = 2.21, p > 0.017.

In piece 2, 26 out of 40 participants chose MMA.  A binomial test revealed that there is not a significant result for either array at a 95% confidence level, z = 1.90, p > 0.017.

B.4.2.3         Clarity

In piece 1, 16 out of 40 participants chose MMA.  A binomial test revealed that there is not a significant result for either array at a 95% confidence level, z = -1.26, p > 0.017.

In piece 2, 9 out of 40 participants chose MMA.  A binomial test revealed that there is a significant result for SFA at a 99% confidence level, z = -3.48, p < 0.003.

B.4.2.4         Naturalness

In piece 1, 16 out of 40 participants chose MMA.  A binomial test revealed that there is not a significant result for either array at a 95% confidence level, z = -1.26, p > 0.017.

In piece 2, 16 out of 40 participants chose MMA.  A binomial test revealed that there is not a significant result for either array at a 95% confidence level, z = -1.26, p > 0.017.

B.4.3      SFI vs. SFA Piece 1 and Piece 2

B.4.3.1         Spaciousness

In piece 1, 13 out of 40 participants chose SFI.  A binomial test revealed that there is not a significant result for either array at a 95% confidence level, z = -2.21, p > 0.017.

In piece 2, 12 out of 40 participants chose SFI.  A binomial test revealed that there is a significant result for SFA at a 95% confidence level, z = -2.53, p < 0.017.

B.4.3.2         Envelopment

In piece 1, 16 out of 40 participants chose SFI.  A binomial test revealed that there is not a significant result for either array at a 95% confidence level, z = -1.26, p > 0.017.

In piece 2, 13 out of 40 participants chose SFI.  A binomial test revealed that there is not a significant result for SFA at a 95% confidence level, z = -2.21, p > 0.017.

B.4.3.3         Clarity

In piece 1, 16 out of 40 participants chose SFI.  A binomial test revealed that there is not a significant result for either array at a 95% confidence level, z = -1.26, p > 0.017.

In piece 2, 19 out of 40 participants chose SFI.  A binomial test revealed that there is not a significant result for either array at a 95% confidence level, z = -0.32, p > 0.017.

B.4.3.4         Naturalness

In piece 1, 19 out of 40 participants chose SFI.  A binomial test revealed that there is not a significant result for either array at a 95% confidence level, z = -0.32, p > 0.017.

In piece 2, 16 out of 40 participants chose SFI.  A binomial test revealed that there is not a significant result for either array at a 95% confidence level, z = -1.26, p > 0.017.

B.5         Attribute Web Test

This section outlines the binomial test results from the attribute web test described in Section 2.3.1.

For the spacious attribute, 17 out of 20 participants chose the more spacious recording extract. A binomial test revealed that this is a significant result at a 99% confidence level, z = 3.13, p < 0.01.

For the envelopment attribute, 18 out of 20 participants chose the more enveloping recording extract. A binomial test revealed that this is a significant result at a 99% confidence level, z = 3.58, p < 0.01.

For the clarity attribute, 19 out of 20 participants chose the more enveloping recording extract. A binomial test revealed that this is a significant result at a 99% confidence level, z = 4.02, p < 0.01.

For the natural attribute, 16 out of 20 participants chose the more natural recording extract. A binomial test revealed that this is a significant result at a 99% confidence level, z = 2.68, p < 0.01.

 

Appendix C.      Saffire Pro 24

This appendix outlines the key specifications of the audio interface used for the playback of the test material. Information is sourced from Focusrite (2014).

C.1         Line level Outputs

• Dynamic Range (A Weighted): 105dB
• SNR (A weighted): 104.5dB
• THD+N: < 0.001% (measured with 0dBFS input and 22Hz/22kHz bandpass filter)
• Maximum level (A weighted): 16.13dBu at 0.885%

C.2         Additional Conversion Performance

• Clock jitter < 250 picoseconds
• THD+N AMPL (A weighted)= 107dBFS

8               Conclusions

Preference results for the Multi Microphone Array vs. the Soundfield ITU definitively favour the Multi Microphone Array. However, it is not reasonable to expect that the recording and mix engineers will not adjust the output settings of the B-format decoding. Soundfield ITU vs. Soundfield Adjusted comparisons, although showing mainly parity, show indications that the adjusted specification gives a better sense of envelopment which promotes the assumption that the recording engineer would adjust default values in search for a better sounding production.

Therefore, the results of the Multi Microphone Array vs. Soundfield Adjusted are more important to consider. For quieter and less energetic pieces, their preference results achieved parity. This suggests that the engineer can focus on what type of sound characteristic they wish to achieve rather than picking an overall preferred array. However, these results could be based out of the spaced vs. coincident Multi Microphone Array considerations whereby the engineer could simply employ a coincident Multi Microphone Array technique to achieve a similar set of image characteristics that the Soundfield gives.

For more energetic pieces, preference was shown to be in favour of the Multi Microphone Array. The Soundfield Adjusted did perform better in the clarity attribute; however, the use of accent microphones or the use of a coincident front Multi Microphone Array can provide reinforcement for the Multi Microphone Array in this aspect. Additionally, an array being clearer is not indicative of positive preference so considerations should be made by the engineer on whether a production which is too clear could irrecoverably damage the perception of the audience.

An approach where clarity is supplemented into the production rather than taken away, which cannot be guaranteed, may be more prudent. Any deficiencies in the Multi Microphone Array used with respect to clarity can be addressed with the variable use of accent microphones or with the use of a coincident Multi Microphone Array for the capture of the front image.

Although a larger sample size would have been beneficial,results of this project show that the Soundfield does not perform significantly better in preference to mean that it should be considered the most competent recording array for classical quartets.

Similarly, any deficiencies in spaciousness or envelopment with the Soundfield system can be addressed with accent microphones; however, this would effectively make redundant the use of the rear pickup of the Soundfield which would not make financial sense. The use of discrete microphones in a Multi Microphone Array also means there is an added versatility with respect to the deployment of microphones across different areas of the performance space as well as in different styles of project.

9               Further Work

Attribute results in this project have shown that preferential results relate to spaciousness, envelopment and clarity. Some questions had results that were close to significance and would have benefitted from a larger sample size. However, enough indications were established to show an avenue for further work in this area. In this case, the borderline parity questions would be asked again if appropriate.

This project compared recording arrays for a classical quartet. This means that results are not directly applicable to a larger ensemble as the need accent microphones would make the assessment of the main array problematic. Additionally, care should be taken when applying these results to other musical sources such as a choir. With these warnings established, further work could explore a similar project question for configurations of other acoustic musical groups in difference spaces.

As outlined in Section 8, the performance of the Soundfield system in terms of clarity may be influenced in the coincident nature of the sound pickup rather than the specific use of the system. Further work in this area can investigate how a surround sound array with a coincident front and spaced rear pickup compares with the Soundfield system or how a coincident stereo array compares with the Soundfield in a stereo configuration.

This project compared arrays in a very favourable performance and recording environment. Queen’s University Belfast provides extremely high quality facilities in these respects. Another avenue for further work would be to recording in substandard or problematic performance spaces to investigate whether the Soundfield system’s post recording versatility translates into advantageous problem solving abilities.

The results presented in the previous section are discussed below, in Sections 7.1 and 7.2. Table 7.1 to Table 7.5 provides a quick reference summary of the results presented in Section 6.1 and 6.2. Highly significant results will be denoted with an asterisk in the summary tables, where appropriate.

Certain considerations should be made when analysing the results. Limitations of the test sample and listening room are discussed in Sections 7.3 and 7.4. Musical considerations are outlined in Section 7.5. Finally, comments made by participants about the test material is highlighted in Section 7.6

7.1           Preference

7.1.1       Overall

	MMA vs. SFI	MMA vs. SFA	SFI vs. SFA
Significant?	Yes*, MMA	No	Yes, SFA

Table 7.1 – Overall Preference

Results show a highly significant preference towards MMA when compared with SFI and parity in preference when MMA is compared with SFA. A significant preference was determined for SFA when compared with SFI. These results will be broken down on a piece by piece basis in the next section.

7.1.2       Piece 1 and Piece 2

	MMA vs. SFI	MMA vs. SFA	SFI vs. SFA
Significant? (Piece 1)	Yes*, MMA	No	No
Significant? (Piece 2)	Yes*, MMA	Yes, MMA	No

Table 7.2 – Piece 1 and 2 Preference

Results for MMA vs. SFI are consistent for each piece at a highly significant level in favour of MMA. With respect to MMA vs. SFA, the second piece shows a distinct jump in votes in favour of MMA to reach significance which indicates that the more energetic music material of the second piece has an impact on array preference. SFI vs. SFA are consistent in parity for both pieces though it should be noted that SFA was at most three votes from a significant result.

Although a single vote from significance, the overall preference results for the MMA vs. SFA back up the findings of Sungyoung et al. (2006) where the Fukada Tree tested was preferred over the Soundfield system; however, by breaking the results into each piece it can be seen that the performance characteristics have a significant impact on participant reactions.

These results are the opposite to the findings of Paquier and Koehl (2011) as their results showed that a naïve group of participants preferred the Soundfield system. However, as described in Section 2.1.3.1, the use of the Big Band would have overloaded the recording arrays. Given this, the clarity displayed by the Soundfield system in this study may explain the preference results of Paquier and Koehl (2011) in the naïve group.

7.2           Attributes

Attribute results are compiled on a piece by piece basis below. For reference, preference results are also provided. An array must achieve a result of at least 28 votes to be significant and 30 votes to be considered highly significant.

7.2.1       Multi Microphone Array vs. Soundfield ITU

	Preference	Space	Envelop	Clarity	Natural
Significant? (Piece 1)	Yes*, MMA	Yes, MMA	Yes*, MMA	Yes, SFI	No
Significant? (Piece 2)	Yes*, MMA	No	Yes, MMA	No	No

Table 7.3 – Piece 1 and 2 Attributes

In piece 1, the MMA was determined more spacious to a significant level; however, in piece 2, parity was achieved. It should be noted that the result is very close to being significant as the votes of an extra participant would have been enough to achieve this.

Results indicate that envelopment and spaciousness contribute to array preference as each attribute received a significant, highly significant or near significant result in favour of the preferred array in piece 1 and 2. Clarity results also indicate that a clearer sounding recording does not translate into overall preference.

7.2.2       Multi Microphone Array vs. Soundfield Adjusted

	Preference	Space	Envelop	Clarity	Natural
Significant? (Piece 1)	No	No	No	No	No
Significant? (Piece 2)	Yes, MMA	No	No	Yes*, SFA	No

Table 7.4 – Piece 1 and 2 Attributes

All attribute results in piece 1 result in parity though it should be noted that the envelopment result is just one vote from significance. In piece 2, clarity is highly significant in favour of SFA which shows that clarity is not a prerequisite to preference.

Interestingly, both arrays achieved a similar number of votes for spaciousness with the envelopment scores being close to significance for MMA. Indications here point towards that each array could deliver the sense of space as well as each other with the ability or inability to envelop a listener being the main difference between them.

7.2.3       Soundfield ITU vs. Soundfield Adjusted

	Preference	Space	Envelop	Clarity	Natural
Significant? (Piece 1)	No	No	No	No	No
Significant? (Piece 2)	No	Yes, SFA	No	No	No

Table 7.5 – Piece 1 and 2 Attributes

Parity was achieved on all comparisons except for spaciousness in piece 2 where SFA was significantly more spacious. This suggests that spaciousness was more noticeable in the more energetic music.

The SFI versus SFA comparisons shared the same front image with the rear image being the only differentiation between them, as described in Section 3.5.1. Given this, the parity results for clarity and naturalness could be influenced by the perception of the front image by participants.

With three of the four combined spacious and envelopment results being borderline between parity and significance, the influence of the rear image is indicated as being very important in the performance of an array with respect to preference. A larger test sample is likely to have benefitted these comparisons.

7.3           Test Sample Limitations

7.3.1       Sample Makeup

Where some significant and highly significant results were established it should be appreciated that the test sample was entirely comprised of people with a background and/or an education in audio. A substantial number of naïve participants could not be sourced due to the time constrains with respect to the booking of Studio D over the main listening room in the University of Salford.

Resulting from this, focus was switched to sourcing participants with a background in audio to make the best use of resources. Although the test sample could be split into naïve and expert listeners with respect to what aspect of audio they are involved in, it would have resulted in two small subdivisions of the test sample which would yield insignificant results.

Only one female volunteered to take part in the listening tests meaning that any gender based considerations were impossible to make. Similarly, the spread of participant age in the test sample also did not result in any age based considerations being possible to make.

7.3.2       Sample Size

Many of the results which were rated as parity would only have required one or two more votes to achieve significance. This translates to the votes of one extra participant. Where preference results were the main aim of the project, more definitive attribute results would have provided a better basis for further research into how the attributes contributed to the preference result. In many cases, indications based on the sample size limitations could only be speculated.

7.4           Listening Room Limitations

The University of Salford listening room, which fully complies with ITU-R BS.1116-1, was not available during the listening test time frame of February 2014 to March 2014. The use of Studio D was the best available space. Section 4.2.1 outlines the compliance of Studio D with ITU-R BS.1116-1 room standard and any areas of non-compliance should be taken into account when considering the results.

7.5           Musical Considerations

This project focused on the recording of a classical quartet, therefore any extrapolations on the suitability of the recording arrays used in this project for the recording of other types of music must be taken with care. An aim of this project is to add to the body of knowledge in this subject area. Further work can add to this knowledge by testing with different musical ensembles and different performance spaces.

The project greatly benefitted from the services of a professional string quartet. This meant that the resulting musical extracts were of a high quality. Additionally, the offer and use of the Queen’s University Belfast microphone and recording equipment meant that the recording signal chain was also of a very high quality, as was the Harty Room performance space. All these aspects must be considered in any future work as a weak element in these areas cannot be fixed without re-recording.

7.6           Participant Comments

Participants were offered an opportunity to give further information on their thoughts and opinions about each pairwise comparison in their own words. Although this information cannot be seen as conclusive and cannot be used as the basis of project conclusion, the comments do correlate to the preference and attribute results while also offering an interesting insight into participant’s thoughts.

Some participants noted that the Multi Microphone Array felt more natural, spacious and warm sounding while at times lacking instrumentation detail. Some felt the Soundfield Adjusted extracts sounded thin and harsh with others feeling that this aspect allowed them to pick apart instruments more clearly but the latter did not necessarily translate into preference for the Soundfield based extract. One participant felt that the Soundfield Adjusted extract felt heavily processed compared with the multi microphone extract. Some participants also felt that the front to rear correlation was too high in the Soundfield extracts.

Next post

8 & 9      Conclusions & Further Work.

This section presents a graphical representation of the binomial test data which is provided in full in Appendix Sections B.2, B.3 and B.4. Results of the preference section are divided into overall and piece by piece results and are presented in Section 6.1. Attribute results are presented on a piece by piece basis in Section 6.2.

All tests were carried out to a confidence level of 95% and where appropriate, 99%. Results which reject the null hypothesis at the 95% confidence level will be described with the respect to their ‘significance’ or as ‘significant’. Results which reject the null hypothesis at 99% will be described with respect to their ‘high significance’ or as ‘highly significant’ (Sprent, 1989, p. 22).

6.1           Preference

6.1.1       Overall

Overall preference results for all test stimuli, across both types of musical pieces are presented. In order to be considered significant, a result of at least 51 votes is required. Results of 54 or more will be considered highly significant.

Figure 6.1 – Overall Preference

6.1.2       Piece 1 and Piece 2

Preference results for each piece of music are presented for comparison against each other. In order to be considered significant, a result of at least 28 votes is required. Results of 30 or more will be considered highly significant.

Figure 6.2 – Piece 1 Preference

Figure 6.3 – Piece 2 Preferences

6.2           Attributes

Attribute results for each piece of music are presented for each pairwise comparison of recording extracts, MMA vs. SFI, MMA vs. SFA and SFI vs. SFA. In order to be considered significant, a result of at 28 or more votes is required where a result of 30 or more will be considered highly significant. For reference, appropriate preference results are also provided.

6.2.1       Multi Microphone Array vs. Soundfield ITU

Results of the comparison between the Multi Microphone Array and Soundfield array in ITU specification for each musical piece are presented.

Figure 6.4 – MMA vs. SFI Piece 1

Figure 6.5 – MMA vs. SFI Piece 1

6.2.2       Multi Microphone Array vs. Soundfield SFA

Results of the comparison between the Multi Microphone Array and Soundfield array in the adjusted specification for each musical piece are presented.

Figure 6.6 – MMA vs. SFA Piece 1

Figure 6.7 – MMA vs. SFA Piece 2

6.2.3       Soundfield ITU vs. Soundfield SFA

Results of the comparison between the Soundfield array in ITU specification and Soundfield in the adjusted specification for each musical piece are presented.

Figure 6.8 – SFI vs. SFA Piece 1

Figure 6.9 – SFI vs. SFA Piece 2

Next post

7       Discussion

5.1           Test Hypotheses

The possible outcomes of the pairwise comparison tests and their relevant hypotheses include:

• Preference towards Multi Microphone Array technique;

o   Test participants may prefer the sound of the Multi Microphone Array recording meaning that despite the post-mix benefits and simplicity of setup associated with the Soundfield microphone, it is not the correct choice in terms of sound quality and listener enjoyment of classical music material. This produces the test hypothesis, H_M, which states that the probability of participants choosing the Multi Microphone Array over the Soundfield will be greater than half.

Equation 5.1 – Hypothesis M

• Preference towards the Soundfield technique;

o   Test participants may prefer the sound of the Soundfield recording meaning that as well as having operational benefits for the engineer, there are also clear sonic advantages to using this microphone for the recording of classical music. This produces the test hypothesis, H_S, which states that the probability of participants choosing the Multi Microphone Array over the Soundfield will be less than half.

Equation 5.2 – Hypothesis S

• Equal preference;

o   Overall results may show parity between each technique. This could mean that rather than having a clearly better tool to record with, the engineer can make a decision on array choice without worrying about negative sonic implications. This produces the test hypothesis, H_E, which states that the probability of participants choosing the Multi Microphone Array over the Soundfield will be half.

Equation 5.3 – Hypothesis E

Given the significant operational differences outlined in Section 1.4, it is unlikely that there will be equal preference, or parity, between a Multi Microphone Array and Soundfield array. Therefore H_M, H_Sand H_E are replaced with H₀ and H_A which are the null and alternative hypotheses respectively as seen in Equation 5.4. These hypotheseswill also be applied to the attribute questions which are detailed in Section 2.3.

Equation 5.4 – Null and Alternative Hypotheses

5.2           Hypothesis Testing

The null hypothesis, H₀, will be tested to establish whether the test question achieves parity. For example, in a situation where a pairwise preference comparison is presented to participants and preference for each is equal, H₀ will not be rejected meaning that equal preference, or parity has been established. In the situation where one of the test options is never selected by the participants, H₀ will be rejected (Sprent, 1989, pp. 7 – 12).

In a case of null hypothesis rejection, the alternative hypothesiswill be established meaning that parity between the pairwise comparisons options has been discounted as a significant difference between the options has been detected to a certain confidence level.To establish where the preference lies between the two options being tested, a two tailed binomial test is used (Bower, 2014; Elder Laboratory, 2014).

If the null hypothesis can not be rejected, the result will have to reside in the blue shaded region which is between -1.96 and 1.96 as shown in Figure 5.1. If the null hypothesis can be rejected, the numerical result will give a figure less than -1.96 or greater than 1.96, depending on the pairwise option which shows significance.

Figure 5.1 – 95% Confidence Interval (e-Discovery Team, 2014)

This numerical region is dependent on the confidence level being used. To determine whether a result is statistically significant, results are commonly compared to a 95% and/or 99% confidence level (Sprent, 1989, p. 8).

To calculate the region, the NORM.S.INV function of Microsoft Excel can be used (Microsoft, 2014). To operate at 99% significance, the critical region is from -2.58 to 2.58. This means that there are more possible answers to the binomial test which can be determined as not rejecting H₀ while results which do reject it, can be held to a higher statistical significance.

5.3           Analysis Formulae

All calculations were completed using two tailed binomial tests, shown in Equation 5.5 where p is the hypothesised proportion, n is the sample size and x is the number of votes of preference for the first option of the test (Bower, 2014; Elder Laboratory, 2014).

Equation 5.5 – One Proportion Binomial Test

Calculations were made with respect to a confidence level of 95% and where appropriate, 99%. As the distribution criteria outlined in Equation 5.6 were satisfied, the normal approximation of the binomial distribution was used (University, West Chester, 2014).

 

Equation 5.6 – Distribution Criteria

As stated in Section 4.1 and further discussed in Appendix B.1, a Bonferroni correction of three was applied to the statistical analysis as three comparisons were being made from each set of recording extract stimuli (Goldman, 2014). This adjusts the critical regions which become -2.39 to 2.39 for 95% and -2.94 to 2.94 for 99%.

Required formulae were integrated into the data collection spread sheet of the project. The integrity of the calculations was tested by using the statistical software package, Minitab (Minitab, 2014).

5.4           Summary

The possible results of the array comparisons have been discussed and appropriate test hypotheses are highlighted. With consideration given to the project aims, an appropriate hypothesis testing method has been established and described. The one proportion binomial test will be used to test the null and alternative hypotheses have which are H₀ and H_A respectively.

Next post

6       Results

4.1           Array Comparison

The three recording arrays being compared are made up of the Multi Microphone Array and two Soundfield derivations which were discussed in Sections 3.3 and 3.5.1 respectively. This section outlines how these arrays were compared. The Multi Microphone Array will be referred to as MMA while the Soundfield in ITU specification and Soundfield in adjusted specification will be referred to as SFI and SFA respectively.

The arrays will be compared in pairwise comparison where arrays will be presented in pairs and participants will be asked questions based on each pair. To test all possible combinations of each array when grouped into pairs, the following pairwise comparisons will be made:

1) MMA vs. SFI

• The Multi Microphone Array versus the Soundfield derivation in ITU specification will highlight if the default settings of the Harpex-B decoding is preferable.

2) MMA vs. SFA

• The Multi Microphone Array versus the adjusted Soundfield derivation will highlight if the post recording benefits of the Soundfield system translates into significant preference for it.

3) SFI vs. SFA

• By comparing the two Soundfield derivations, differences between them can be established. As the front of the image for both derivations is the same, results here will focus on how the difference in rear pickup will influence preference.

The results of the comparisons one and two can have a bearing on the results of comparison three. For example, if the MMA is preferred in comparison one and SFA is preferred in comparison two, then SFI should not be preferred in comparison three as shown in Equation 4.1.

Equation 4.1 – Expected Results

Two pieces of music were selected for testing. Each differed in performance characteristics such as perceived loudness and dynamic range. Two twenty second sections of each piece were selected for use in the test. Each section was sourced from distinct parts of the musical piece. The aim of using these was to establish whether the difference in performance characteristics had an influence on the results. Table 4.1 outlines the comparison of the arrays used. As there are three comparisons per set, a Bonferroni correction of three was be applied to the statistical analysis which is detailed in full in Appendix B.1.

Piece 1, section 1	Piece 1, section 2	Piece 2, section 1	Piece 2, section 2
MMA vs. SFI	MMA vs. SFI	MMA vs. SFI	MMA vs. SFI
MMA vs. SFA	MMA vs. SFA	MMA vs. SFA	MMA vs. SFA
SFI vs. SFA	SFI vs. SFA	SFI vs. SFA	SFI vs. SFA

Table 4.1 – Extract Comparison

4.2           The Listening Room

Studio D of the Newton building is the University of Salford’s audio post production and mastering suite. Equipment applicable to this project includes:

• Blue Sky Sat 6.5 MkII surround monitoring system including LFE.
• Digidesign C24 control surface with 5.1 monitor controller.

Studio D is isolated from the building structure and other studios by way of acoustic damping to help eliminate external noise influencing the environment within the studio. It has been designed to minimise the effects of room modes and is acoustically treated to provide an accurate monitoring environment (Salford University, 2014). The studio’s dimensions are shown in Figure 4.1.

Figure 4.1 – Studio D dimensions (H. Mattsson, personal communication, 15 April 2014)

4.2.1       ITU-R BS.1116-1 Room Standards

This section outlines the room requirements with respect to background noise, room dimensions and speaker setup given by the ITU-R BS.1116-1 subjective audio testing standard and outlines the compliance of Studio D in these respects.

4.2.1.1      Area

The room floor area for multichannel playback situations should be between 30m² and 70m². The floor area of Studio D is 15.86m² so does not meet this requirement (ITU, BS.1116-1, 1997, p. 10).

4.2.1.2      Shape

The listening room should be vertically symmetrical in the centre of the room, perpendicular to the listening position which studio D complies with (ITU, BS.1116-1, 1997, p. 10). The room should also be in a rectangle or trapezoid shape (ITU, BS.1116-1, 1997, p. 10). Studio D complies with the former; however, the corners on the listening end are tapered to facilitate a door way in one corner and room symmetry in the other. These characteristics are shown in Figure 4.1.

4.2.1.3      Proportions

ITU-R BS.1116-1 supplies three formulae to assess the suitability of the room’s dimensions (ITU, BS.1116-1, 1997, p. 11). Studio D fails the first, as can be seen in Equation 4.2; however, it satisfies the remaining equations which are detailed in Equation 4.3 and Equation 4.4.

Equation 4.2 – BS.1116-1 Room Dimensions Calculation and Result 1

Equation 4.3 – BS.1116-1 Room Dimensions Calculation and Result 2

Equation 4.4 – BS.1116-1 Room Dimensions Calculation and Result 3

4.2.1.4      Background Noise

Figure 4.2 – Background Noise Measurement

The background noise of the testing venue is recommended to be below the NR10 noise rating curve (ITU, BS.1116-1, 1997, p. 14). Figure 4.2 shows the actual reading versus the ITU recommendations. The measurements were made by configuring Studio D for the listening test and switching off all unrequired equipment.

4.2.1.5      Listening Level

The ITU-R BS.1116-1 calculations state that for playback over a five speaker system, playback levels were to be calibrated to 83.6dBA, as shown in Equation 4.5 (ITU, BS.1116-1, 1997, p. 14). However, playback level was calibrated to 80dBA as per guidelines of the testing venue and with considerations of the findings by the European scientific committee on Emerging and Newly Identified Health Risks (European Commission, 2014). It should be noted that this level adjustment is within the margin described in Section 4.3.4 which outlines the facility for participants to adjust the playback level to a comfortable setting within a 6dB margin above and below the ITU-R BS.1116-1 calculated listening level.

 

Equation 4.5 – BS.1116-1 Playback Level Calculation and Result

4.2.1.6      Monitor Placement

The height of monitors should be 1.2 metres high and at least 1 metre from each wall (ITU, BS.1116-1, 1997, p. 15). Studio D meets the former; however, all speakers are within a metre of the wall due to space constraints which is reflected by the room area calculations shown in Section 4.2.1.1.

4.2.1.7      Listener Position

The distance from the loudspeaker to the listener should be between 2 and 5.1 metres and meet ITU-R BS.775-3 speaker angle specifications (ITU, BS.1116-1, 1997, p. 16). Studio D meets these criteria.

4.3           Delivery

A total of twenty-one volunteers took part in the listening tests. The results of one participant were excluded in post screening as they skipped two sections of the test. Out of the remaining allowable results, participant ages ranged from nineteen to thirty-eight years with an average age of twenty-four years. Twenty males and one female took part. All participants were students at the University of Salford. Subject areas ranged from undergraduate video, audio and acoustic courses to postgraduate audio, acoustics and spatial audio courses.

Participants took part in the listening test individually. Each test took place in Studio D in the Newton Building of the University of Salford. This studio was the most appropriate listening space available during the project time frame to meet the ITU-R BS.1116-1. A Max 6.1 patch was used to play the audio material, collect answers and export them for statistical analysis. A Dell E6530 was used to run the test interface, provide keyboard and mouse input and playback the audio extracts. A Focusrite Saffire Pro 24 firewire audio interface was used to patch into Studio D monitoring equipment via the patch bay. Technical specifications of the Pro 24 interface can be found in Appendix C.

4.3.1       Introduction and Briefing

Participants used the provided keyboard and mouse to fill details such as their age, gender and their occupation. These details would be used to identify any age, gender or occupation based trends providing the test sample offers such opportunities.

Participants were then briefed in what to expect from the test in relation to duration and questioning. An introductory paragraph of text was provided on screen which welcomes them to the session and describes the testing process. They were introduced to the test interface and guided through the training and interface familiarisation section.

4.3.2       Participant Training

With respect to the attribute questions of the test, each definition was supported by two audio extracts which displayed more of and less of each attribute. Participants were then asked to select which extract displayed more of each definition. Once the participant was comfortable in the aspects of the test, the test supervisor left the room. The supervisor used a window, behind the participant, to monitor how they interacted with the interface. If an issue was identified, the supervisor would enter the room to correct it.

4.3.3       Test Interface

Cycling74’s Max 6.1 software was used to create the test interface (Cycling 74, 2014). It is a piece of object based programming software designed for multimedia and interactive applications. It was used to:

• Provide a means of training of participants,
• Provide a test interface which will allow for pairwise comparison of two recording extracts,
• Provide a means to randomise playback of test stimuli,
• Provide looped surround sound playback of test stimuli,
• Provide real time switching of test stimuli,
• Allow for participants to submit their answers to test questions,
• Provide test data for collation and processing

Details of the patch can be found in Appendix A.

A copy of the pairwise comparison sections was provided to participants, as seen in Figure 4.3. In this section, participants were given details on each part of the test interface. Participants were encouraged to press various buttons to familiarise themselves with the progression of the test. For example, when all ‘Click me!’ buttons were pressed and the answers were confirmed, the patch displayed a box instructing the participant to move to the next section.

With the participant familiar and comfortable with the interface and text questions, they continued to the twelve pairwise comparison sections. All recording extracts used in the test up until this point were derived from a separate performance from the same recording session from what was used in the paired comparison sections. The definitions were supplied to participants on screen and on paper for reference throughout the test.

Figure 4.3 – Example of the layout section

4.3.4       Listening Level Adjustment

If required, the listening level of the playback system could be adjusted to a comfortable level by the listener. The ITU-R BS.1116-1 documentation outlines a reference listening level calculation for testing playback which is outlined Section 4.2.1.5 which also details the alternative use of a maximum level of 80dBA.

Given that the general impressions made by musical sound sources were being assessed rather than defects in audio codecs which ITU-R BS.1116-1 covers, a flexibility of ± 6dBA of the listening level was given to the participants to allow for comfortable listening levels of the dynamic music (ITU, BS.1116-1, 1997, pp. 12, 13). The adjusted playback level of 80dBA outlined in Section 4.2.1.5 is within the ± 6dBA margin and although this facility of level adjustment was offered to participants during each test, all participants were comfortable with the initial playback level of 80dBA.

4.3.5       Completion

With the test complete, the participants were instructed by the test interface program to contact the test supervisor. Participants were thanked for their time with any of their questions answered and requested not to share any information about the test with any waiting participants. Their results were then logged and saved in a master Excel spread sheet.

4.4           Summary

The makeup of the test sample has been described. The testing delivery method and procedure is outlined. The Multi Microphone Array and two derivations of the Soundfield system were compared together using a non-parametric pairwise comparison method. The results given by participants were categorical with respect to their preference between the presented pair of stimuli and for which stimulus in a pair conveyed the most of certain sonic attributes. The testing venue was compliant in most respects of the relevant testing standards.

Next post

5       Statistical Analysis