| Data compression plays an important role in modern telecommunication, entertainment, computing and networking areas because the capacities of media for storage and bandwidth for transmission are not growing proportionally with the rapid demand for multimedia data. Speech and voice compression is one of applications in this field. Although there are a lot of techniques used in speech coding, new algorithms need to be developed to achieve better performance. Our research focus is on lossless speech and voice compression using wavelet transform, prediction, and Rice coding. These techniques have some properties, such as, very fast computation and easily exploiting the redundancy in the speech signal. By taking these advantages, we can reduce the number of bits required to represent the audio signal and get better lossless audio compression. In this thesis, basic concepts and properties of these techniques will be displayed and a new loseless algorithm based on these methods is presented as well. The test results of the new algorithm are shown and some analyses are given. |
