Sign in

username:

password:



Not a member?

Search Online Books



Search tips

Free Online Books

Ads

Chapters

Physical Audio Signal Processing
    Artificial Reverberation
       The Reverberation Problem
          Exact Reverb via Transfer-Function Modeling

Chapter Contents:

Search Physical Audio Signal Processing

  

Book Index | Global Index

Would you like to be notified by email when Julius Orion Smith III publishes a new entry into his blog?

  

Exact Reverb via Transfer-Function Modeling

Figure 2.1 depicts the general reverberation scenario for three sources and one listener (two ears). In general, the filters should also include filtering by the pinnae of the ears, so that each echo can be perceived as coming from the correct angle of arrival in 3D space; in other words, at least some reverberant reflections should be spatialized so that they appear to come from their natural directions in 3D space [254]. Again, the filters change if anything changes in the listening space, including source or listener position. All prevalent artificial reverberation systems implement some approximation of the system in Fig.2.1.

Figure 2.1: General reverberation simulation for three sources and one listener (two ears). Each filter $ h_{ij}$ can be implemented as a tapped delay line FIR filter.
\begin{figure}\input fig/reverb.pstex_t \end{figure}

In the frequency domain, it is convenient to express the input-output relationship in terms of the transfer-function matrix:

$\displaystyle \left[\begin{array}{c} Y_1(z) \\ [2pt] Y_2(z) \end{array}\right] ... ...left[\begin{array}{c} S_1(z) \\ [2pt] S_2(z) \\ [2pt] S_3(z)\end{array}\right] $

Denoting the impulse response of the filter from source $ j$ to ear $ i$ by $ h_{ij}(n)$, the two output signals in Fig.2.1 are computed by six convolutions:

$\displaystyle y_i(n) = \sum_{j=1}^3 s_j \ast h_{ij}(n) = \sum_{j=1}^3 \sum_{m=0}^{M_{ij}} s_j(m)h_{ij}(n-m), \quad i=1,2, $

where $ M_{ij}$ denotes the order of FIR filter $ h_{ij}$. Since many of the filter coefficients $ h_{ij}(n)$ are zero (at least for small $ n$), it is more efficient to implement them as tapped delay lines (§1.5) so that the inner sum becomes sparse. For greater accuracy, each tap may include a lowpass filter which models air absorption [322] and/or spherical spreading loss (see §1.3). For large $ n$, the impulse responses are not sparse, and we must either implement very expensive FIR filters, or approximate the tail of the impulse response using less expensive