State of open-source software for microphone array processing

Oliver Ackermann Lylloff, Gert Herold, Adam Kujawski, Ennes Sarradj

Research output: Contribution to conferencePaperResearchpeer-review

Abstract

Open-source software has become increasingly important in research and recognized through funding and peer-reviewing. This paper outlines the current state of openly available software for microphone array processing and describes two libraries in detail: the well-established Acoular, and newer addition AeroAcoustics.jl. For both libraries, the design choices as well as possible application is discussed. Moreover, it is detailed how these open-source libraries allow a multi-leveled approach to quality assurance through the possibility of software peer review, regression tests as well as open benchmarks. The latter is also shown to be instrumental for reproducible research.
This is further exemplified by demonstrating a benchmark exercise with measured data from a wind tunnel test. The task is performed with both libraries. The presented results highlight the possibility to get the same results from the same data using different software.
Finally, lessons learned from the development of these tools are discussed, i.e., funding, peer-reviewing, student and researcher collaboration between research teams, and community involvement. It is concluded that open-source scientific software can be a key for the scientific community to improve not only the quality and reproducibility of research and application, but also the general availability of new algorithms and results.
Original languageEnglish
Publication date2024
Number of pages14
Publication statusPublished - 2024
Event10th Berlin Beamforming Conference 2024 - Berlin, Germany
Duration: 10 Jun 202411 Jun 2024

Conference

Conference10th Berlin Beamforming Conference 2024
Country/TerritoryGermany
CityBerlin
Period10/06/202411/06/2024

Fingerprint

Dive into the research topics of 'State of open-source software for microphone array processing'. Together they form a unique fingerprint.

Cite this