Nowadays the measurement technology of room-acoustics could apply tools with very high computing capaticy. With these it is possible to solve almost real-time signal processing problems, in other words we can evaluate the measured data right on the analysed location.
If we choose the signal path properly, it’s also possible to improve some previously acquisited local acoustical parameters manually (by moving or changing acoustic surfaces) or automatically (with the algorithmic alternation of the to-be-outputted sound samples)
The second solution seemed realisable, or at least seemed worthy to start dealing with as a thesis subject.
The final aim is to produce an application that can detect and eliminate objectively describable sound errors. With this feature it can help to learn previously unknown acoustical environments faster, for example for the own sound crew of a touring band in foreign concert halls or musicians touring without sound personnel.
This thesis-plan documents the first development step of this plan: the creating of a room acoustics measurement tool for concert halls using a PA-system. The tool is planned to be able to generate and emit measuring signals through the PA, record the room-modified version of them, and process the measured data on the spot.
I had to choose between two different ways of the implementation: designing and building a dedicated standalone hardware or writing a software for a hardware system using more universal components (such as a portable computer and soundcard). I began to work on the second option, because it has more future advantages: for example when we decide to implement new features on the system we don’t need to purchase new dedicated parts. The other great advantage is that this way a personal computer does the computing tasks (for example signal processing or real-time compensation). The personal computer technology is still in the fast evolving status, so a measurement system based on it can operate faster from year to year.
My work consists of distinct sections: first I had to recall my studies in programming, signal processing, room acoustics and measurement technology. After that I connected my chosen development toolkit (MATLAB 7.5) to the external soundcard’s input and output channels. Then I implemented the measurement software concentrating on easy operation and low system resource usage. Up to now I haven’t achieved to develop the real-time correction function.
As most of the room acoustic parameters are derived from the impulse response(s) of a room, my program is based on measuring that function. Measuring can be executed with different inducing signals: we can choose between noises with wide frequency range, pieces of music or swept-frequency signals. Besides the software is able to analyse the time-delay caused by the electronic signal path and sound propagation.
As a final accomplishment I compared my system with an acoustics measurement device composed by Péter Fiala, professor of the Department of Telecommunications. Although his device is more sophisticated and multifunctional, with the same measure arrangements the two devices measured very similar transfer functions.