标题: 3D声音讯号处理实现于汽车音讯:以单麦克风背景噪音估测为基础之适应性音讯增益控制器
3D Audio Signal Processing on Automotive Audio : Adaptive Audio Gain Controller Based on One-Microphone Noise Level Estimation
作者: 艾学安
白明宪
机械工程学系
关键字: 虚拟重低音;语音清晰化;动态控制系统;讯噪比;静态曲线;Virtual bass;Voice clarity;Dynamic range control;Signal to noise ratio;Static curve
公开日期: 2008
摘要: 本论文的研究着重于提升汽车音讯的聆听品质。现今车内的多声道视听系统进展的十分快速,然而,汽车内部仍然因为一些背景噪音像引擎、震动、冷气而被视为不良的聆听环境。再来叙述一些可以在车内恶劣环境下拓展频宽以及提升聆听品质的声音系统。本研究提出三个方法在车内产生声音特效,第一、二部份分别叙述虚拟重低音(VB)以及语音清晰化(VC)的方法,它们分别阐述将频宽往低频拓展以及强调中频的技术。第三部份提出声道拓展/压缩,提出如何针对狭小且反射多的环境提供较好的声场。本研究也提出了一个新的系统,运用噪音估测系统(NLE)结合动态控制系统(DRC)来适应性的调整汽车音讯系统的增益。此方法运用了两个系统,首先使用最小均方(LMS)演算法来估测背景噪音的大小,之后计算先验的讯噪比(SNR)再根据事先设计的静态曲线(Static curve)可以得到适应性调整后的增益。
A comprehensive study was conducted to improve the listening quality of automotive audio. There is increased proliferation nowadays of multichannel audiovisual systems used in cars. However, the interior of a car is known as a notorious listening environment due to background noise like engine, vibration and air conditioning. It is then desirable to develop audio systems that are capable of extending the bandwidth and improving the audio quality in harsh car environments. This study brings three approaches to generate the audio effect in a car. First and second section is virtual bass (VB) and voice clarity (VC), describes the bandwidth expansion toward the low-frequency and emphasizes the middle-frequency. Third section is updownmix, describes how to render spatial sound field to cope with the reflections in the confined space. This study also proposes a new system that makes use of a Noise Level Estimator (NLE) combines Dynamic Range Control (DRC) to adaptively adjust the gain of an automotive audio system. NLE and DRC are used in the system for noise estimation and automatic gain control, respectively. The system needs one microphone to receive the noisy signal. Background noise level is estimated by a system that uses Least-Mean-Squares (LMS) algorithm and then calculated a priori signal to noise ratio (SNR). According to the static curve designed in advance, the gain of the audio input can be adjusted dynamically based on the SNR determined. These processing algorithms have been practically implemented on a car. Simulations and experiments were conducted for validating the proposed adaptive audio gain control systems.
URI: http://140.113.39.130/cdrfb3/record/nctu/#GT079614576
http://hdl.handle.net/11536/42142
显示于类别:Thesis


文件中的档案:

  1. 457601.pdf

If it is a zip file, please download the file and unzip it, then open index.html in a browser to view the full text content.