Abstract—In this paper is to evaluate audio and speech quality with the help of Digital Audio Watermarking Technique under the different types of attacks (signal impairments) like Gaussian Noise, Compression Error and Jittering Effect. Further attacks are considered as Hostile Environment. Audio and Speech Quality Evaluation is an important research topic. The traditional way for speech quality evaluation is using subjective tests. They are reliable, but very expensive, time consuming, and cannot be used in certain applications such as online monitoring. Objective models, based on human perception, were developed to predict the results of subjective tests. The existing objective methods require either the original speech or complicated computation model, which makes some applications of quality evaluation impossible. Keywords—Digital Watermarking, DCT, Speech Quality, Attacks. I. INTRODUCTION ITH the rapid growth of network distributions of digital media contents, there is an urgent need for copyright protection against piracy. The embedding of digital watermarks into multimedia content has been proposed to tackle this kind of problem. Audio watermarks are special information signals embedded into digital audio. These signals are extracted by detection mechanisms. Robustness and imperceptibility are important requirements of watermarking. The evaluation of audio and speech quality is a very important field in current multimedia era. According to specific practice of long standing, the only way to measure the quality of an audio signal was through the use of subjective quality evaluation [1].In this test ten or more person were involved to listen a live or recorded conversation and assign a rating to it. Participants listened to the audio sequences and were asked to report using five-point scale: (5: imperceptible, 4: perceptible but not annoying, 3: slightly annoying, 2: annoying, 1: very annoying). The arithmetic mean of the collection of these was taken for quality of audio which is called the mean opinion score (MOS). This has been the most reliable method of speech quality assessment but it is highly unsuitable for online Akhil Kumar Arya was M.Tech Student with the Electrical Engineering Department, National Institute of Technology Kurukshetra Haryana India (e- mail:akhilarya@gmail.com) Jagdeep Singh Lather, is with the Electrical Engineering Department, National Institute of Technology Kurukshetra Haryana India (e- mail:jslather@nitkkr.ac.in) Dr.Lillie Dewan is with the Electrical Engineering Department, National Institute of Technology Kurukshetra Haryana India (e-mail: l_dewan@nitkkr.ac.in) monitoring applications and is also very expensive and time consuming. Due to these reasons, objective methods have been developed in recent years, classified into two categories: signal-based methods and parameters- based methods [5]. The signal-based methods use the reference and degraded signals as the input to the measurement; on the contrary, the parameters-based methods predict the speech quality through a computational model instead of using real measurement. Objective methods can also be classified as intrusive and nonintrusive ones. Intrusive method takes both the original and the degraded speeches as the input. Non-intrusive methods only require the degraded speech. It is more challenging to design a nonintrusive method because no original speech information could be used during the quality evaluation. Recently, several nonintrusive speech quality evaluation methods have been proposed. In the following, we will briefly introduce subjective, signal based objective and parameters-based objective speech quality evaluation methods. 1) Signal Based Methods: Signal-based methods use the reference and distorted signals as input. The two signals are compared based on some perceptual model and the predictions of subjective test results are generated. In order to achieve an estimate of perceived quality, a measurement should employ as much understanding of human perception and human judgments as possible. The common idea behind perceptual quality measurement is to mimic the situation of a subjective test [1]. 2)Parameter Based Methods: Besides perceptual measurement, some other parameters based methods, such as Gaussian mixture models, artificial neural networks and E- models, have also been developed for audio and speech quality assessment. In Gaussian mixture models (GMMs), a large pool of feature measurements is extracted and created from the distortion surface between the original speech signal and the degraded speech signal. Good features are then chosen [4]. The joint density of these selected features is modeled with the subjective MOS as a Gaussian mixture. Finally, using this model, the least squares estimate of the subjective MOS value is derived. This model outperforms the PESQ in root mean square errors but the improvement in correlation between the subjective MOS and predicted MOS is small. In, artificial neural networks (ANNs) have been employed to assess audio quality in packet networks with the concern of several distortion parameters on transmitted audio, such as, arrival jitter, end-to-end delay, sampling rate and the number A Tool for Audio Quality Evaluation Under Hostile Environment Akhil Kumar Arya, Jagdeep Singh Lather,Lillie Dewan W World Academy of Science, Engineering and Technology International Journal of Electrical and Computer Engineering Vol:2, No:3, 2008 487 International Scholarly and Scientific Research & Innovation 2(3) 2008 ISNI:0000000091950263 Open Science Index, Electrical and Computer Engineering Vol:2, No:3, 2008 publications.waset.org/12615/pdf