AES Los Angeles 2014
Poster Session P13
P13 - Applications in Audio: Part 1
Saturday, October 11, 10:30 am — 12:00 pm (S-Foyer 1)
P13-1 Automated Sound Optimization of Car Audio Systems Using Binaural Measurements and Parametric IIR Filters—Friedrich von Tuerckheim, Visteon Electronics Germany GmbH; Tobias Münch, Visteon Electronics Germany GmbH - Karlsruhe, Germany
Sound tuning is an important step towards improved listening conditions in car interiors. In most cases it is done manually by sound engineers. This paper presents an approach for fully automated sound optimization. In a first step, loudspeaker and interior responses are captured by averaged binaural measurements. Then, the resulting frequency response is matched to a given reference curve. As automotive head units often provide limited capacity for audio filters, a small set of second order recursive filters is used for equalization. Numerical optimization leads to a minimum error response while maintaining psychoacoustic specifications. The presented method is used for fast and efficient frequency response correction as well as for copying sound characteristics of different car interiors.
Convention Paper 9166
P13-2 Study of TV Sound Level Adjustment System for the Elderly with Speech Rate Conversion Function—Tomoyasu Komori, NHK Engineering System, Inc. - Setagaya-ku, Tokyo, Japan; Waseda University - Shinjuku-ku, Tokyo, Japan; Atsushi Imai, NHK Science & Technology Research Laboratories - Setagaya-ku, Tokyo, Japan; Nobumasa Seiyama, NHK Science & Technology Research Laboratories - Setagaya-ku, Tokyo, Japan; Reiko Takou, NHK Science & Technology Research Laboratories - Setagaya-ku, Tokyo, Japan; Tohru Takagi, NHK Science & Technology Research Laboratories - Setagaya-ku, Tokyo, Japan; Yasuhiro Oikawa, Waseda University - Shinjuku-ku, Tokyo, Japan
Elderly viewers sometimes feel that background sound (music and sound effects) in TV programs is too loud, or that narration or speech is too fast to understand. That is why we have constructed a prototype system that compensates for both of these problems with sound on the receiver side. The results of evaluation experiments targeting elderly viewers showed that the use of this system could make it significantly easier to listen to TV sound. These results also showed that elderly viewers exhibit the "recruitment phenomenon." They tend to select processing with a slowed speech rate that is easy to hear.
Convention Paper 9167
P13-3 Investigation of Gain Adjustment in a Personal Assistive Listening System Using Parametric Array Loudspeakers—Santi Peksi, Nanyang Technological University - Singapore; Woon-Seng Gan, Nanyang Technological University - Singapore, Singapore; Ee-Leng Tan, Nanyang Technological University - Singapore, Singapore; Eu-Chin Ho, Tan Tock Seng Hospital - Singapore; Satya Vijay Reddy Medapati, Tan Tock Seng Hospital - Singapore
Human hearing degrades with ages, which leads to difficulties in viewers of different age groups enjoying television together as they required different audio volumes. To address the problem Simon et al.  proposed loudspeaker arrays that boost 10 dB at all frequencies in a narrow spatial zone where hearing-impaired listener is located. This paper presents a different approach using a personal assistive listening (PAL) system that aims to deliver a highly directional sound beam with the required gain amplification through a parametric array loudspeaker to match the hearing profile of a hearing-impaired listener, while delivering normal sound loudness to the rest of normal listeners using conventional electro-dynamic loudspeakers. This paper investigates the gain adjustment of two commercially-available parametric loudspeakers over the frequency range for audiometry testing and relates the gain adjustments to the sound pressure level (SPL) at various positions away from the sound system.
Convention Paper 9168
P13-4 Cinema Sound Facility Design for Higher Education—Robert Jay Ellis-Geiger, City University of Hong Kong - Hong Kong, SAR China
This paper is a narrative of the trials and tribulations that the author went through from design through to the commissioning of probably the most advanced higher education cinema sound facilities within the Asia-Pacific region. The facilities include a 7.1 THX and Dolby certified dubbing theatre, audio recording studio integrated into a 30-workstation audio/music technology lab, multiple 5.1 surround screening rooms, color correction, multi-format home entertainment environment and a large sound stage that can accommodate a full symphonic orchestra. The main purpose for the facilities were to support the delivery of undergraduate and post-graduate courses in sound, music, and audio within the academic studios of cinematic arts and animation and to establish a research center for cinema sound and music technology applications.
Convention Paper 9169
P13-5 A General-Purpose Decorrelator Algorithm with Transient Fidelity—Ross Penniman, University of Miami - Coral Gables, FL, USA
In a multichannel spatial audio presentation, a decorrelator is a signal-processing algorithm that helps to create a diffuse sensation for the listener by defeating any localization cues. In this paper the relevant psychoacoustic and signal processing principles are reviewed, and a new decorrelator algorithm is proposed that operates blindly on a single-channel input signal and creates a 5-channel decorrelated presentation. This algorithm uses transient extraction to achieve better fidelity when decorrelating a wide range of input signals. A subjective listening test compares the performance of the proposed algorithm in relation to two existing algorithms drawn from the literature. Results of the test are discussed as well as suggested improvements to the test methodology.
Convention Paper 9170
P13-6 Applicability of Perceptual Evaluation of Speech Quality in Evaluating Heavily Distorted Speech—Mitsunori Mizumachi, Kyushu Institute of Technology - Kitakyushu, Fukuoka, Japan
Speech quality assessment is indispensable to properly design a speech enhancement algorithm. The perceptual evaluation of speech quality (PESQ) is frequently employed as an objective speech distortion measure. The PESQ is a methodology for estimating subjective assessment of speech quality assuming a slight distortion caused by speech codecs for telephony systems. In case of noise reduction, however, a degree of speech distortion is heavier than those caused by the speech codecs. In this paper applicability of the PESQ is investigated for noisy and noise-reduced speech signals under severe noisy conditions. A relationship between PESQ scores and subjective mean opinion scores reveals that the PESQ can be applicable for heavily distorted speech only under non-stationary noisy conditions.
Convention Paper 9171