Style transfer" among images has recently emerged as a very active research topic, fuelled by the power of convolution neural networks (CNNs), and has become fast a very popular technology in social media. This paper investigates the analogous problem in the audio domain: How to transfer the style of a reference audio signal to a target audio content?We propose a flexible framework for the task, which uses a sound texture model to extract statistics characterizing the reference audio style, followed by an optimization-based audio texture synthesis to modify the target content. In contrast to mainstream optimization-based visual transfer method, the proposed process is initialized by the target content instead of random noise and the optimized loss is only about texture, not structure. These differences proved key for audio style transfer in our experiments. In order to extract features of interest, we investigate different architectures, whether pretrained on other tasks, as done in image style transfer, or engineered based on the human auditory system. Experimental results on different types of audio signal confirm the potential of the proposed approach.Index Terms-Audio style transfer, sound texture model, texture synthesis, deep neural network, auditory system.

Despite having conceptual and practical advantages, Complex-Valued Neural Networkss (CVNNs) have been much less explored for audio signal processing tasks than their real-valued counterparts. We investigate the use of a complex-valued Convolutional Recurrent Neural Network (CRNN) for Direction-of-Arrival (DOA) estimation of a single sound source on an enclosed room. By training and testing our model with recordings from the DCASE 2019 dataset, we show our architecture compares favourably to a real-valued CRNN counterpart both in terms of estimation error as well as speed of convergence. We also show visualizations of the complex-valued feature representations learned by our method and provide interpretations for them.

In many signal processing applications, metadata may be advantageously used in conjunction with a high dimensional signal to produce a desired output. In the case of classical Sound Source Localization (SSL) algorithms, information from a high dimensional, multichannel audio signals received by many distributed microphones is combined with information describing acoustic properties of the scene, such as the microphones' coordinates in space, to estimate the position of a sound source. We introduce Dual Input Neural Networks (DI-NNs) as a simple and effective way to model these two data types in a neural network. We train and evaluate our proposed DI-NN on scenarios of varying difficulty and realism and compare it against an alternative architecture, a classical Least-Squares (LS) method as well as a classical Convolutional Recurrent Neural Network (CRNN). Our results show that the DI-NN significantly outperforms the baselines, achieving a five times lower localization error than the LS method and two times lower than the CRNN in a test dataset of real recordings.