Sökning: L773:1558 7916 OR L773:1558 7924 >
Optimizing the simi...
Optimizing the similarity of loudspeaker : Room responses in multiple listening positions
-
- Bahne, Adrian (författare)
- Uppsala universitet,Signaler och System
-
- Ahlén, Anders (författare)
- Uppsala universitet,Signaler och System
-
(creator_code:org_t)
- 2016
- 2016
- Engelska.
-
Ingår i: IEEE Transactions on Audio, Speech, and Language Processing. - 1558-7916 .- 1558-7924. ; 24:2, s. 340-353
- Relaterad länk:
-
https://urn.kb.se/re...
-
visa fler...
-
https://doi.org/10.1...
-
visa färre...
Abstract
Ämnesord
Stäng
- A shortcoming of multichannel sound reproduction standards, such as stereo or 5.1 surround, is their incompatibility with multiple off-axis listening positions. Accurate reproduction of virtual sound sources can only be experienced in the sweet spot, which is located equidistant to the loudspeakers. We here present a novel methodology to compensate audio systems such that the channel similarity is optimized in several listening positions simultaneously. To that end we propose a novel MIMO personal audio filter design framework based on feed-forward control. By proper design choices, filters that successfully compensate for multiple offaxis positions and irregularities in the frequency sum responses are obtained. The design choices include allpass filters with appropriate phase shifts as target for each listening position in addition to a weighted similarity requirement. Evaluations based on measurements of two four-channel car audio systems show that the proposed method significantly improves timbral sound reproduction and phantom center reproduction in several listening positions simultaneously.
Nyckelord
- Acoustic signal processing; audio systems; equalizers; loudspeakers; multiple sweet spots; psychoacoustics; virtual center
- Elektroteknik med inriktning mot signalbehandling
- Electrical Engineering with specialization in Signal Processing
Publikations- och innehållstyp
- ref (ämneskategori)
- art (ämneskategori)
Hitta via bibliotek
Till lärosätets databas