IoSR Blog : 27 November 2015
AES New York Convention Report
The most recent Audio Engineering Society convention was held in New York from October 29th to November 1st 2015. Three IoSR researchers were in attendance and presented four very well received papers.
Jon Francombe presented two papers on research undertaken as part of the EPSRC-funded S3A project. The first paper [1] considered loudness measurement for 3D audio. Loudness measurement is necessary in broadcast and audio research; however, until recently, standardised models had not been tested with modern spatial audio systems. The paper described a perceptual experiment in which listeners matched the loudness of a range of audio programmes for different reproduction methods. These results were then used to test a number of existing loudness models. The results from this paper contributed to international standardisation efforts for measuring loudness of advanced spatial audio systems. The second paper [2] considered various techniques for visualising multichannel room impulse response datasets, which are used in various aspects of spatial audio including room acoustics, object capture, and reverberation.
Kirsten Hermes presented a paper about listening test methodology [3]. The topic of the paper was 'dumping bias', a source of bias that occurs when participants are asked to rate changes in a single attribute (in this case, 'clarity’), but their responses are skewed because changes in other attributes are also present. The research showed that requiring participants to rate multiple attributes—either simultaneously to or before the attribute of interest—reduces statistical noise in listening test results.
Andy Pearce presented early research from his PhD (supported by Cirrus Logic), describing methods that can be used to capture recordings in order to compare microphones [4]. The research also experimentally showed the maximum array spacing that can be used to make recordings for microphone comparisons, as well as demonstrating a test methodology that can be used to find a maximum array size for any recording environment.
As well as these (and many other) papers, there was a host of interesting workshops, tutorials, and demonstrations. For example, there was a tutorial on current microphone design and specifications, which gave an overview of the topic that gave an opportunity to see how professional microphone designers interpret the objective measures of microphones - a subject closely related to Andy’s work.
There were many sessions focusing on surround sound with a height element. These included papers that reported investigations of recording techniques and the psychoacoustics of with-height surround sound, as well as demonstrations of immersive spatial sound (including MPEG-H audio). It was good to see the research community working on methods for capturing and reproducing surround sound with height, as this has been investigated in the S3A project (e.g. a paper in which various recordings were made in different surround sound formats was presented by Jon at the previous AES convention [5]).
The theme of game audio also seemed to be fairly prevalent at the convention. Many different facets were covered, from innovations in game audio technology and audio immersion, to presentations on compositional considerations for aspiring musicians. An interesting talk was given by Winifred Phillips, one of the composers of the music from the Little Big Planet series of games. She discussed the novel take this game had on audio. The score for each level comprises six audio layers, each of which must be able to be played alone and still hold as a solid backing music, yet also to integrate with all other layers simultaneously and sound like a cohesive piece of music.
It was useful to be able to discuss research with other researchers but also industry practitioners; Kirsten spoke to mix engineers Buford Jones and George Massenburg about the parameters of high quality music mixes, giving an interesting insight into how award-winning practitioners might answer some of her research questions from a creative point of view. The sheer size of the convention, and the range of the sessions available, helps to facilitate this kind of cross-discipline dialogue. Of course, New York was also a fantastic venue; on one night, the Empire State building was even lit up in AES colours in celebration of the convention!
- 1. J.Francombe, T.Brookes, R.Mason, F.Melchior, "Loudness Matching Multichannel Audio Program Material with Listeners and Predictive Models", Audio Eng. Soc. 139th International Convention, New York, October 2015
- 2. L.Remaggi, P.Jackson, P.Coleman, J.Francombe, "Visualization of Compact Microphone Array Room Impulse Responses", Audio Eng. Soc. 139th International Convention, New York, October 2015
- 3. K.Hermes, T.Brookes, C.Hummersone, "The Influence of Dumping Bias on Timbral Clarity Ratings", Audio Eng. Soc. 139th International Convention, New York, October 2015
- 4. A.Pearce, T.Brookes, M.Dewhirst, "Validation of Experimental Methods to Record Stimuli for Microphone Comparisons", Audio Eng. Soc. 139th International Convention, New York, October 2015
- [5] Francombe, J., Brookes, T., Mason, R., Flindt, R., Coleman, P., Liu, Q., and Jackson, P.J.B., 2015: “Production and reproduction of programme material for a variety of spatial audio formats”, Audio Engineering Society 138th Convention, eBrief No. 199, Warsaw, Poland, 7-10 May 2015.
by Jon Francombe, Kirsten Hermes, and Andy Pearce