AES London 2011
Paper Session P20
P20 - Subjective Evaluation
Monday, May 16, 14:00 — 16:00 (Room 1)
P20-1 Selection of Audio Stimuli for Listening Tests—Jonas Ekeroot, Jan Berg, Arne Nykänen, Luleå University of Technology - Luleå, Sweden
Two listening test methods in common use for the subjective assessment of audio quality are the ITU-R recommendations BS.1116-1 for small impairments and BS.1534-1 (MUSHRA) for intermediate quality. They stipulate the usage of only critical audio stimuli (BS.1116-1) to reveal differences among systems under test, or critical audio stimuli that represents typical audio material in a specific application context (MUSHRA). A poor selection of stimuli can cause experimental insensitivity and introduce bias, leading to inconclusive results. At the same time this selection process is time-consuming and labor-intensive, and is difficult to conduct in a systematic way. This paper reviews and discusses the selection of audio stimuli in listening test-related studies.
Convention Paper 8445 (Purchase now)
P20-2 A Listening Test System for Automotive Audio: PART 5 – The Influence of Listening Environment on the Realism of Binaural Reproduction—Francois Postel, Bang & Olufsen A/S - Struer, Denmark (now at Arkamys, Paris, France), Department of Mechanical Vibrations and Acoustics, UTC, Compiegne, France; Patrick Hegarty, Søren Beck, Bang & Olufsen A/S - Struer, Denmark
Binaural technology is used to capture elements of an in-car sound field and reproduce them over headphones at another place and time. An experiment to test the influence of the listening environment on the realism of such a binaural reproduction is described. A panel of 12 trained listeners rated a range of stimuli for 6 elicited attributes of sound quality. Ratings are made for the actual sound field in the test vehicle, for a binaural reproduction in the same vehicle, and for a binaural reproduction in a listening room. The results show that the tested binaural reproduction system is able to preserve either the rank order or the perceived magnitudes of the impressions of the sound field for the attributes Precision, Treble, Stereo impression, Bass, and Reverberation, independent of the listening environment.
Convention Paper 8446 (Purchase now)
P20-3 Differences in Preference for Noise Reduction Strength between Individual Listeners—Rolph Houben, Academic Medical Center - Amsterdam, The Netherlands; Tjeerd M. H. Dijkstra, Radboud University - Nijmegen, The Netherlands; Wouter A. Dreschler, Academic Medical Center - Amsterdam, The Netherlands
There is little research on user preference for different settings of noise reduction, especially for individual users. We therefore measured individual preference for pairs of audio streams differing in noise reduction strength. Data was analyzed with a logistic probability model that is based on a quadratic preference utility function. This allowed for an estimate of the optimal setting for each individual subject. For five out of ten subjects the optimized setting differed significantly from the optimum obtained for the grouped data. However, the predicted preference for the individual optimum (60%) was only slightly higher than chance level (50%), which can be considered as too weak to advocate individualization of noise reduction for these normally hearing subjects. However, in hearing-impaired subjects this may be different.
Convention Paper 8447 (Purchase now)
P20-4 Assessment of Stereo to Surround Upmixers for Broadcasting—David Marston, BBC R&D - London, UK
Broadcasters are now transmitting 5.1-channel surround sound as part of their HD TV services. However, since much of the audio content in original program material is 2-channel stereo, broadcasters are required to switch between the two formats. Switching between 2- and 5.1-channel formats can cause problems in decoding the content, including switching artifacts and loudness changes. In general it is preferable to transmit all program audio in 5.1 surround, and this can be achieved by automatically upmixing any 2-channel stereo content to 5.1 format prior to broadcasting. This paper reports on tests designed to assess the subjective performance of a selection of upmixers for use in the broadcast chain.
Convention Paper 8448 (Purchase now)