Lossless audio coding fully preserves the quality of music using only a fraction of the size
MSc (Tech) Florin Ghido developed improved methods for lossless coding of audio signals, which provide smaller compressed files and faster decoding speeds than the existing lossless audio coding standards and formats, such as MPEG-4 ALS, MPEG-4 SLS, and FLAC.Lossless compression exploits the redundancy present in the input data in order to achieve a more compact representation, while ensuring bit-identical reconstruction. The best known example is the ZIP format, used to reduce the size of computer data files.
For music in the Audio-CD format, ZIP programs can only compress the data to around 92 % of the original size, whereas specialized programs, such as those developed in the thesis, achieve a typical size reduction of 2:1 (to 50 %) up to 4:1 (to 25 %), depending on the type of content. In contrast, lossy audio coding, such as MP3 and AAC, achieves a significantly higher size reduction, in the range of 4:1 to 11:1. These formats additionally discard fine details present in the data, which are generally not considered audible by humans. However, in particular listening conditions some of the missing details or even the introduced coding artifacts may become detectable.
The thesis contains theoretical contributions which provide a better understanding and modeling of the properties of audio signals with respect to efficient coding and, at a practical level, it proposes flexible and configurable algorithmic structures, including a complete codec design.
Lossless audio coding has applications for multimedia libraries, archival of music collections, multimedia distribution between recording studios, live broadcasts for news channels, online music stores, and portable multimedia players, such as music players, smartphones, and tablets. Lossless audio coding is also widely used in DVD-Audio and Blu-ray discs, where it enables storage of high-resolution multichannel audio (e.g., using 96 kHz, 24 bit, 8 independent channels).
Public defence of a doctoral dissertation on Friday, 15 February
The doctoral dissertation of MSc (Tech) Florin Ghido in the field of signal processing titled “Modeling and Optimization for Lossless Audio Coding with Flexible Complexity Profiles” will be publicly examined at the Faculty of Computing and Electrical Engineering of Tampere University of Technology (TUT) in room TB109 in the Tietotalo building (address: Korkeakoulunkatu 1, Tampere, Finland) at 12:00 on Friday, 15 February. Dr. Aki Härmä from Philips Research Europe, The Netherlands and Postdoctoral Researcher Heidi-Maria Lehtonen from Aalto University will act as the opponents. Professor Ioan Tabus from the Department of Signal Processing at TUT will act as Chairman.
Further information: Florin Ghido, puh. +358 40 416 1110, florin.ghido@tut.fi
http://www.losslessaudio.org/