We present a method to compute the delay-constrained multicast routing tree by employing chaotic neural networks. The experimental result shows that the noisy chaotic neural network (NCNN) provides an optimal solution more often compared to the transiently chaotic neural network (TCNN) and the Hopfield neural network (HNN). Furthermore, compared with the bounded shortest multicast algorithm (BSMA), the NCNN is able to find multicast trees with lower cost.