| dc.description.abstract | Nowadays the multimedia applications of the 3D virtual reality are more and more popular. Although most application focus on 3D video, the combination of 3D video and audio processing can enrich the experience of users. Most of the traditional stereo systems, which can only maintain partial sound spatial distribution, provide listeners with the ability to distinguish the two-dimensional plane, not the performance of three-dimensional spatial acoustics. Multi-channel surround sound systems come about to solve this problem, but this kind of systems require the sound source to be treated specially, also it has severe restrictions to hardware and environment. All of these restrain its applications. In view of this situation, researchers developed the theory of virtual surround sound (VSS) and carry about many research studies.
The objective of this thesis is to explore and create a dual-channel virtual surround sound system. The thesis first introduces fundamentals of acoustics and human sound location theory, including the cues of precedence effect, Pinna effect, ITD, IID, head related transfer function (HRTF), reflection and reverberation. Then we propose a virtual sound source system using two loudspeakers, which can perform crosstalk cancellation. To confirm the feasibility of this method, we do experiments by computer simulations using MATLAB and obtain satisfactory effects.
We evaluate the system by observing the phase between two channels after the proposed processing and the subjective audio quality of the assessment on the surround effect. In three different types of audio tests, audio surround effect is significantly enhanced. Through simulation of humans spatial hearing characteristics, VSS system can reproduce normal sound source’s 3D surround effect using a pair of loudspeakers. This system has simple configuration and is able to produce lifelike sense. It has great potential to provide extensive applications in small room that is not suitable for multi-channel surround systems. | en_US |