ACTA ACUSTICA UNITED WITH ACUSTICA Vol. 90 (2004) 1178 – 1184 The Influence of the Directivity of Musical Instruments in a Room Felipe Otondo, Jens Holger Rindel Ørsted-DTU, Acoustic Technology, Technical University of Denmark, building 352, Ørsteds Plads, DK-2800, Kgs.Lyngby, Denmark Summary Measurements of the directivity of musical instruments are presented as part of the study of the influence of their representation in room acoustic simulations and auralizations. Pairs of measured and averaged directivities have been used both for room simulation comparisons and as a basis for listening experiments with auralizations. Room simulation results show a clear influence of the changes in the representation directivity on the distribution of acoustical parameters in the room. The results of the listening experiments with auralizations show that some changes produced by directivity variations can be perceived by the listener. Among these changes, loudness was perceived in the highest degree, followed by reverberance and clarity. PACS no. 43.75.-z, 43.55.Ka, 43.55.Hy 1. Introduction The directivity of musical instruments has been studied by several authors [1, 2, 3, 4], J¨ urgen Meyer being the one who has contributed with most specific information about the radiation characteristics of musical instruments in a real performance situation [5]. However, Meyer’s directiv- ity data mostly provide directivity averages over the whole performing frequency range of the instruments. Very little information about the directivity of instruments for par- ticular tones is included, even though it is shown that the directivity can change dramatically over the performing range [5]. Most of the data available nowadays on the di- rectivity of musical instruments used for room acoustic simulations and auralizations use the averaged directivi- ties from Meyer’s studies. A few attempts have been made to study and use a different directional representation, a representation that would include the directivity changes of the musical instruments within the performing range [6, 7, 8, 9]. At the same time, experiments using room acoustics auralizations have shown that the directional rep- resentation of sources in room acoustic simulations is im- portant, and that changes in their directivity can signifi- cantly affect the perceived sound in a room [10]. Therefore, the first goal of this investigation has been to measure the directivity of different musical instruments in a performance situation and compare it to the averaged directivity used traditionally for room simulation and au- ralization purposes. The second goal has been to use these two types of directivity representation in room simulations Received 11 April 2003, accepted 2 February 2004. in order to investigate how important the changes in the ra- diation representation can be in the spatial distribution of acoustical parameters in a room. The third goal has been to use the two types of directivities to build auralizations in order to evaluate the perceptual importance of the changes in the radiation in a room by means of listening experi- ments. 2. Directivity of Musical Instruments 2.1. Directivity measurements Three different classical musical instruments with dis- tinctive directional behaviours were used for the direc- tivity measurements: a Bb clarinet, a Bb trumpet and a French horn. The measurements were carried out as a post- processing of simultaneous recordings done with thirteen flat response electret microphones in an anechoic chamber at 45 intervals in the horizontal and vertical planes, using a 24-bit quantization and 44.1 kHz sampling rate at a dis- tance of 1.5 meters from the source. Figure 1 shows one of the performers playing a trumpet in the anechoic chamber and Figure 2 shows the recording/measuring setup with the thirteen microphones in the horizontal and vertical planes. As it is shown in Figure 2 the setup chosen for the direc- tivity measurements was conceived in such a way as to measure the variations of directivity of the instruments in the horizontal and vertical planes according to the tradi- tional representation of the directivity of sound sources in the room acoustic software available nowadays. The goal with this setup was not to obtain a detailed description of the whole radiation sphere of the instrument, but to ob- tain a sample of the variations of the radiation in both planes to be used afterwards in the room simulations soft- 1178 c S. Hirzel Verlag EAA