Convolutional Neural Networks (CNNs) have had great success in many machine vision as well as machine audition tasks. Many image recognition network architectures have consequently been adapted for audio processing tasks. However, despite some successes, the performance of many of these did not translate from the image to the audio domain. For example, very deep architectures such as ResNet [1] and DenseNet [2], which significantly outperform VGG [3] in image recognition, do not perform better in audio processing tasks such as Acoustic Scene Classification (ASC). In this paper, we investigate the reasons why such powerful architectures perform worse in ASC compared to simpler models (e.g., VGG). To this end, we analyse the receptive field (RF) of these CNNs and demonstrate the importance of the RF to the generalization capability of the models. Using our receptive field analysis, we adapt both ResNet and DenseNet, achieving state-of-theart performance and eventually outperforming the VGG-based models. We introduce systematic ways of adapting the RF in CNNs, and present results on three data sets that show how changing the RF over the time and frequency dimensions affects a model's performance. Our experimental results show that very small or very large RFs can cause performance degradation, but deep models can be made to generalize well by carefully choosing an appropriate RF size within a certain range.

This work addresses the problem of matching musical audio directly to sheet music, without any higherlevel abstract representation. We propose a method that learns joint embedding spaces for short excerpts of audio and their respective counterparts in sheet music images, using multimodal convolutional neural networks. Given the learned representations, we show how to utilize them for two sheet-music-related tasks: (1) piece/score identification from audio queries and (2) retrieving relevant performances given a score as a search query. All retrieval models are trained and evaluated on a new, large scale multimodal audio-sheet music dataset which is made publicly available along with this article. The dataset comprises 479 precisely annotated solo piano pieces by 53 composers, for a total of 1,129 pages of music and about 15 hours of aligned audio, which was synthesized from these scores. Going beyond this synthetic training data, we carry out first retrieval experiments using scans of real sheet music of high complexity (e.g., nearly the complete solo piano works by Frederic Chopin) and commercial recordings by famous concert pianists. Our results suggest that the proposed method, in combination with the large-scale dataset, yields retrieval models that successfully generalize to data way beyond the synthetic training data used for model building.

We introduce Deep Linear Discriminant Analysis (DeepLDA) which learns linearly separable latent representations in an end-to-end fashion. Classic LDA extracts features which preserve class separability and is used for dimensionality reduction for many classification problems. The central idea of this paper is to put LDA on top of a deep neural network. This can be seen as a non-linear extension of classic LDA. Instead of maximizing the likelihood of target labels for individual samples, we propose an objective function that pushes the network to produce feature distributions which: (a) have low variance within the same class and (b) high variance between different classes. Our objective is derived from the general LDA eigenvalue problem and still allows to train with stochastic gradient descent and back-propagation. For evaluation we test our approach on three different benchmark datasets (MNIST, CIFAR-10 and STL-10). DeepLDA produces competitive results on MNIST and CIFAR-10 and outperforms a network trained with categorical cross entropy (having the same architecture) on a supervised setting of STL-10.

No abstract

Cross-modality retrieval encompasses retrieval tasks where the fetched items are of a different type than the search query, e.g., retrieving pictures relevant to a given text query. The state-of-the-art approach to cross-modality retrieval relies on learning a joint embedding space of the two modalities, where items from either modality are retrieved using nearest-neighbor search. In this work, we introduce a neural network layer based on canonical correlation analysis (CCA) that learns better embedding spaces by analytically computing projections that maximize correlation. In contrast to previous approaches, the CCA layer allows us to combine existing objectives for embedding space learning, such as pairwise ranking losses, with the optimal projections of CCA. We show the effectiveness of our approach for cross-modality retrieval on three different scenarios (text-to-image, audio-sheet-music and zero-shot retrieval), surpassing both Deep CCA and a multi-view network using freely learned projections optimized by a pairwise ranking loss, especially when little training data is available (the code for all three methods is released at: https://github.com/CPJKU/cca_layer).