v7.0, 20040922, me
Friday, October 29, 9:00 am 11:00 am
Session Z2 Posters: INSTRUMENTATION AND MEASUREMENT & LOSSY AND LOSSLESS AUDIO CODING
NOTE: During the first 10 minutes of the session all authors will present a brief outline of their presentation.
Z2-1 A New Digital Measurement for Distortion of Acoustical DevicesKeiichi Imaoka, Juro Ohga, Shibaura Institute of Technology, Minato-ku, Tokyo, Japan
There is still no suitable measuring method by a digital processing system for nonlinear distortion of acoustical devices. This paper presents a new amplitude nonlinearity measurement by using a Pink-TSP (time stretched pulse) signal. This method applies a TSP signal, whose frequency band is partially eliminated, to the device. The detected component produced in the rejected band is measured as a distortion.
Convention Paper 6216
Z2-2 Measurement of Small-Size Loudspeaker Units by New Acoustical LoadsYusuke Nakano, Juro Ohga, Shibaura Institute of Technology, Minato-ku City, Tokyo, Japan
This paper describes a new measuring method for small sized loudspeakers by using a tube load. The acoustical loads defined in IEC standards for loudspeaker measurement, both of closed boxes and a baffle, are larger in size than the practical acoustical loads for small loudspeakers, for example, mobile telephone bodies. This paper proposes a tube load for measurement and examines practical methods without any effect by tube resonance.
Convention Paper 6217
Z2-3 Extending Quasi-Anechoic Measurements to Low FrequenciesEric Benjamin, Dolby Laboratories, San Francisco, CA, USA
It is often desirable to make electro-acoustic measurements in ordinary working spaces. These measurements would normally be performed in anechoic chambers. Various techniques have been evolved to make what are commonly referred to as quasi-anechoic measurements. These techniques make use of the fact that the initial signal from a loudspeaker-microphone system is anechoic, until the first reflection arrives. By analyzing only that portion of the signal which arrives before the first echo, an anechoic measurement is achieved. These measurements have a low-frequency limitation due to the shortness of the reflection-free time window. Time-frequency tradeoffs in the transformation of the Impulse Response to the frequency domain make it difficult to accurately estimate of the response of the device under test. We first characterize the nature of the errors induced by the short time window and then propose a methodology for reducing the error.
Convention Paper 6218
Z2-4 Efficient AAC Single Layer TranscoderChun-Yi Lee, Cheng-Han Yang, Te-Hsueh Lai, Tihao Chiang, Hsueh-Ming Hang, National Chiao Tung University, Hsinchu, Taiwan
This paper presents a novel algorithm for transcoding the MPEG-4 AAC single-layer bit-streams for bit-rate adaptation purposes. The delivery of multimedia over heterogeneous networks and to the devices with varying capabilities calls for the bit-rate adaptation capability. A previous approach that cascades a pair of full-grown decoder and encoder has very high computational complexity. Our approach can reduce the complexity drastically; however, its coding performance is close to that of the previous cascaded method. In order to achieve this simplification goal, three rate-distortion models/techniques have been employed.
Convention Paper 6219
Z2-5 Effective Tonality Detection Algorithm Based on Spectrum Energy in Perceptual Audio CoderKeun-Sup Lee, Yonsei University, Seoul, Korea; Kyu-Chel Yeon, LG Electronics Inc., Seoul, Korea; Young-Cheol Park, Yonsei University, Seoul, Korea; Dae Hee Youn, Yonsei University, Seoul, Korea
The goal of the perceptual audio coder is to reduce redundancy and irrelevancy of audio signals based on the concept of masking. Several studies on the masking effect reveal that the masking threshold varies as a function of the noise-like or tone-like nature of audio signals. Therefore, the tonality of audio signals influence significantly the quality and efficiency of the perceptual audio coder. In this paper, we proposed a new effective algorithm for tonality measurements using spectrum energy. The performance of the proposed algorithm is comparable to the MPEG audio psychoacoustic model II (PAM-II). However, since the proposed algorithm consists of simple operations plus a few transcendental functions, computational complexity is much lower than the PAM-II. The proposed algorithm was tested with audio signals. DSP implementation showed that the proposed algorithm could be implemented with 2.88 MIPS.
Convention Paper 6220
Z2-6 Audio Patch Method in MPEG-4 HE AAC DecoderHan-Wen Hsu, Chi-Min Liu, Wen-Chieh Lee, National Chiao Tung University, Hsin-Chu, Taiwan; Zheng-Wen Li, InterVideo Digital Technology (Shanghai) Co., ShangHai, China
This paper extends the previous work on AAC to the HE-AAC. The audio path method consists of two individual parts, zero band dithering and high frequency reconstruction. The zero band dithering can conceal the fishy artifact in the low frequency part that is encoded by a convention AAC encoder. Furthermore, high frequency reconstruction can extend the audio obtained from the SBR to a full bandwidth signal. Intensive experiments have been conducted on various encoders and audio tracks to check the quality improvement and the possible risks in degrading the quality. The objective test measures used is the recommendation system by ITU-R Task Group 10/4.
Convention Paper 6221
Z2-7 Bit-Weighted Inter-Channel Prediction for Subband Audio CodingCheng-Han Yang, Hsueh-Ming Hang, National Chiao Tung University, Hsinchu, Taiwan
An efficient algorithm for removing interchannel redundancy in subband audio coding is presented in this paper. In our approach, the bit-weighted interchannel prediction is applied to the Modified Discrete Cosine Transform (MDCT) coefficients. Similar to the INT-DCT based approach, no audio quality degradation will be induced by our method. In addition, the bit rate reduction performance of our method is about 8 percent better than that of the INT-DCT based approach for the cases that interchannel prediction is useful.
Convention Paper 6222