Simulative investigation of required spatial source resolution in directional room impulse response measurements

International audienceIn classical and standardized room acoustic measurements, the directivity of sources and receivers is either not considered, specified to be omnidirectional for inter-measurement comparability, or for the receiver defined as HRTF or spatial audio formats. These room impulse response (RIR) measurements allow for the analysis of certain aspects of an acoustic scene. Parameters such as Early Decay Time (EDT) attempt to reflect the listening impression, while others such as the reverberation time serve as physical descriptors and for the evaluation of rooms for different purposes. The explicit non-consideration of the source and receiver directivity waives an abundance of information, which is essential for a complete description of the spatial composition of an acoustic scene, for example in an auralization. On the one hand, a more complete description raises the degree of realism in auralizations, presumably intensifying the immersion of human listeners. On the other hand, the freedom of arbitrary source-receiver directivity combinations opens up the possibility of directional acoustical analysis like scanning single reflection paths within a room.For the measurement of directional room impulse responses (DRIR) several measurement methods and instruments have been implemented in the past. Most of them aim either at fast measurements or at measurements of a high spatial resolution. The compromise of obtaining a sufficiently high resolution in an acceptable time usually is often disregarded due to the ambiguous definition of a sufficiently high resolution.Before the development of suitable methods and instruments, the importance of directivity in an acoustical scene has to be determined. This entails the question if and up to which complexity the measurement of directional room impulse responses offers an advantage for room representations in auralizations and for parametric room acoustic descriptions.A sufficiently high resolution can be defined in many ways, either in regard to just noticeable differences for human listeners, or objectively regarding the effects on technical parameters. This work investigates the effect of the source directivity resolution on room acoustical parameters. Since real measurements contain too many uncontrollable influences such as time variances and sources that can radiate the required directivity with a sufficient precision do not yet exist, this investigation is done using room acoustic simulations.First, the modelling of suitable artificial directivity will be explained. The spatial resolution will be denoted by the corresponding spherical harmonic order. The goal is a directivity with a minimum beam width for each given spherical harmonic order without strong side lobes. This characteristic represents the worst case for each resolution. The generated source directivity will then be used in hybrid ray tracing and image source room acoustic simulations in three rooms of different size and acoustic property. This approach allows a more generally valid statement about the impact of the source directivity.The results will be discussed using the impact on objective room acoustic parameters as an indicator for the required spatial source resolution. Subjective parameters will be considered as an outlook towards the impact on the human perception. The findings are meant to aid the design of measurement instruments for directional room impulse response measurement in reasonable measurement times with a sufficiently high spatial resolution