Examensarbeit

Perceptual audio coding techniques, such as mp3 and AAC, are well known and deployed on many portable devices, like smartphones, portable music players etc., since they allow for efficient transmission or storage of large media collections. Semantic audio coding technologies enable the media consumer to interact with the audio content while listening to it, for instance, to change the volume and even the position of individual instruments in an audio mix for creating a personalized listening experience. The listener can, for example, remove the vocals or single instruments from the mixture and play or sing along instead. In sports event broadcasts, the user can adjust the level of the announcers or crowd noises individually according to his or her likings.

Fraunhofer IIS and the International Audio Laboratories Erlangen have openings in the development of an advanced semantic audio coding technology called Spatial Audio Object Coding (ISO/MPEG SAOC). Various positions are available, such as student internships, student jobs, or bachelor and master/diploma theses. The work tasks primarily focus on the optimization of individual operational building blocks, e.g. time-frequency transforms, filter banks, content analysis and parametrization, and efficient bit stream coding techniques. The tasks also encompass the subsequent evaluation of the work, employing objective and subjective criteria, by performing measurements and conducting listening tests using real-world signals.

Vorausgesetzte Vorlesungen bzw. Kenntnisse:

Applicants are expected to have a good knowledge of digital signal processing theory, a basic knowledge of Matlab and matrix calculus, and, most importantly, interest and dedication in developing advanced multimedia processing as part of a team of innovative researchers.

Schlagwörter:

Audio, Music, Signal Processing, Fourier Transform, MP3

Bearbeitungszustand:

Die Arbeit ist noch offen.



	UnivIS ist ein Produkt der Config eG, Buckenhof