Colloquium - Details
You will receive information about presentations in time if you subscribe to the newsletter of the Colloquium Communications Technology.
Master-Vortrag: Primary-Ambient Aware Sound Field Translation
Irina Kharlamova
Mittwoch, 21. April 2021
11:00 Uhr
virtueller Konferenzraum
Sound field translation is a technique that allows the sound field to be represented in a displaced coordinate system. This master's thesis presents a coherence-adaptive method of sound field translation based on space warping techniques. The proposed method moves mainly only the direct part of the signal and keeps the ambient part relatively unchanged. the system first identifies the direct and ambient components of the signal and then uses this data to build a filter-mask. For the first task, the method uses parametric decomposition of the sound field. A series of experiments have been conducted with single and multiple sources presented as directional noise or speech, also including reverberation. In experiments comparing other different translation methods, the proposed coherence-adaptive approach achieves excellent performance in terms of various instrumental metrics.
