Fig. 11From: Synthesis of soundfields through irregular loudspeaker arrays based on convolutional neural networksAmplitude (real part) of the soundfield for a source placed in \(\textbf{r}=[-3.76, \text {m}, -1.14\ \text {m}, 0\ \text {m}]^T\) at \(f=1500\ \text {Hz}\), ground truth is shown in a. Reproduction performances using the irregular circular array of \(L=32\) using \(\text {MR}\)Â (b), \(\text {CNN}\)Â (c), \(\text {PM}\)Â (d), \(\text {AWFS}\)Â (e), and \(\text {APWD}\)Â (f)Back to article page