When two computer systems communicate, for example over a network, they must agree on the ordering of bytes within numbers. This ordering is called endianess. Often one of the systems has to swap the order of bytes to the agreed standard. Results of this work help programmers to find places in their program where they forgot to swap the bytes. We have developed a dynamic data-flow analysis built upon the popular Valgrind tool. Compared to the static analysis currently used by the Linux kernel developers, our approach does not require annotation of variables with their endianess. Typically only few places in the program source code will need to be annotated. The analysis can also detect potential bugs that would only manifest if the program was run on computer with opposite endianess. Our approach has been validated on an existing program known to contain yet unfixed endianess problems (RadeonSI OpenGL driver). It has identified all endianess-related bugs and provided useful diagnostic messages together with their location.

In recent years there has been an increasing interest in dynamic taint tracing of compiled software as a powerful analysis method for security and other purposes. Most existing approaches are highly application specific and tends to sacrifice precision in favor of performance. In this thesis project a generic taint tracing tool has been developed that can deliver high precision taint information. By allowing an arbitrary number of taint labels to be stored for every tainted byte, accurate taint propagation can be achieved for values that are derived from multiple input bytes. The tool has been developed for x86 Linux systems using the dynamic binary instrumentation framework Valgrind. The basic theory of taint tracing and multi-label taint propagation is discussed, as well as the main concepts of implementing a taint tracing tool using dynamic binary instrumentation. The impact of multi-label taint propagation on performance and precision is evaluated. While multi-label taint propagation has a considerable impact on performance, experiments carried out using the tool show that large amounts of taint information is lost with approximate methods using only one label per tainted byte.

This diploma thesis discusses the issue of Voice Activity Detection. There are two types of detectors described: energetic and statistics. Their funktionality is proved in the Matlab environment. Further, the implemetation of VADs is made through C language with standard libraries and GSL open-source libraries. The realized algorithms are compared in the scope of processing time of computation, memory management and a single mathematical operations stress. Also a comparism of the processing time according to segment length was made.

This thesis deals with description of human hearing, audio and speech codecs, description of objective measure of quality and practical comparison of codecs. Chapter about audio codecs consists of description of lossless codec FLAC and lossy codecs MP3 and Ogg Vorbis. In chapter about speech codecs is description of linear predictive coding and G.729 and OPUS codecs. Evaluation of quality consists of description of segmental signal-to- noise ratio and perceptual evaluation of quality – WSS and PESQ. Last chapter deals with description od practical part of this thesis, that is comparison of memory and time consumption of audio codecs and perceptual evaluation of speech codecs quality.