There is an increasing interest in the automatic digitization of medieval music documents. Despite efforts in this field, the detection of the different layers of information on these documents still poses difficulties. The use of Deep Neural Networks techniques has reported outstanding results in many areas related to computer vision. Consequently, in this paper, we study the so-called Convolutional Neural Networks (CNN) for performing the automatic document processing of music score images. This process is focused on layering the image into its constituent parts (namely, background, staff lines, music notes, and text) by training a classifier with examples of these parts. A comprehensive experimentation in terms of the configuration of the networks was carried out, which illustrates interesting results as regards to both the efficiency and effectiveness of these models. In addition, a cross-manuscript adaptation experiment was presented in which the networks are evaluated on a different manuscript from the one they were trained. The results suggest that the CNN is capable of adapting its knowledge, and so starting from a pre-trained CNN reduces (or eliminates) the need for new labeled data.

Two experiments were conducted to perceptually evaluate a virtual acoustic room model implemented using a virtual microphone technique for the direct path and reflections, the image-source model for the reflection geometry and a Feedback Delay Network reverberator. Experiment 1 aimed to determine the best equalization profile for different room sizes. Anechoic recordings of 7 solo instruments were processed to generate 5 room sizes with profiles corresponding to 5 different absorption settings. Fourteen sound engineers and musicians were asked to rate the resulting stimuli in terms of how realistic each setting was for the instrument in that space. The highest ranked absorption setting for each room size was selected for Experiment 2 to investigate whether listeners could identify different rooms sizes across different instruments. Using a free sorting task, 15 musicians were presented with 7 different solo instruments in 5 different room sizes and asked to group them by type of space in which the instruments were performed. The analysis of the dissimilarity matrices indicates that participants were able to organize sounds by room sizes, but the two largest rooms tended to be clustered together. The results show that the virtual acoustic model is capable of rendering realistic room effects.