The field of Music Emotion Recognition has become and established research sub-domain of Music Information Retrieval. Less attention has been directed towards the counterpart domain of Audio Emotion Recognition, which focuses upon detection of emotional stimuli resulting from non-musical sound. By better understanding how sounds provoke emotional responses in an audience, it may be possible to enhance the work of sound designers. The work in this paper uses the International Affective Digital Sounds set. A total of 76 features are extracted from the sounds, spanning the time and frequency domains. The features are then subjected to an initial analysis to determine what level of similarity exists between pairs of features measured using Pearson's r correlation coefficient before being used as inputs to a multiple regression model to determine their weighting and relative importance. The features are then used as the input to two machine learning approaches: regression modelling and artificial neural networks in order to determine their ability to predict the emotional dimensions of arousal and valence. It was found that a small number of strong correlations exist between the features and that a greater number of features contribute significantly to the predictive power of emotional valence, rather than arousal. Shallow neural networks perform significantly better than a range of regression models and the best performing networks were able to account for 64.4% of the variance in prediction of arousal and 65.4% in the case of valence. These findings are a major improvement over those encountered in the literature. Several extensions of this research are discussed, including work related to improving data sets as well as the modelling processes.

No abstract

No abstract

No abstract

From the visual music films of the twentieth century to the Video Jockey (VJ) performances seen at the latest electronic dance music festivals, there is an extensive body of artistic work that seeks to visualize sound and music. The form that these visualizations take has been shaped significantly by the capabilities of available technologies; thus, we have seen a transition from paint to film; from hand-drawn animations to motion-graphics; and from analog to digital projection systems. In the twenty-first century, visualizations of music are now possible with extended reality (XR) technologies such as virtual reality (VR), augmented/mixed reality (AR/MR), and related forms of multi-projection environment such as fulldome. However, the successful design of visual music and VJ performances using XR technologies requires us to consider the compositional approaches that can be used by artists and designers. To investigate this area, this chapter will begin with an analysis of existing work that visualizes music using XR technologies. This will allow us to consider the spectrum of existing design approaches, and provide a commentary on the possibilities and limitations of the respective technologies. Following this, the chapter will provide an in-depth discussion of Weinel's practice-led research, which extends from work exhibited at the Carbon Meets Silicon exhibitions held at Wrexham Glynd ŵr University (2015University ( , 2017, and includes AR paintings, VJ performances, and a VR application: Cyberdream VR. Through the discussion of these works, the chapter will demonstrate possible compositional principles for visualizing music across media ranging from paint to XR, enabling the realization of work that reinforces the conceptual meanings associated with music.

No abstract

This paper discusses a variety of the author's artistic projects exploring altered states of consciousness and computer art. First, the paper will provide a brief overview of previous creative works, which include compositions of electroacoustic music, interactive visualisations, and visual music films. These previous works use the concept of altered states of consciousness as a compositional principle, as explored in the author's book Inner Sound: Altered States of Consciousness in Electronic Music and AudioVisual Media (OUP 2018). Following this, a variety of the author's recent creative work produced from 2016-2019 will be discussed. These works include: a series of paintings that incorporate computer graphics animations when viewed in augmented reality; VJ performances constructed using direct animation on 8mm film, computer graphics animations generated from code and audio-reactive effects; and Cyberdream VR, a virtual reality experience. These interrelated projects continue to develop the author's artistic investigations into altered states, while also referencing work such as demo scene videos; cyberdelic imagery of the type seen on fliers from the 1990s rave-era; and the recent Internet-borne subculture vaporwave, which recontextualises the aesthetics of 1980s and 1990s ambient corporate music and utopian computer graphics to construct surrealistic dystopias.

In this paper, we undertake an initial study evaluation of a recently developed audio compression approach; Audio Compression Exploiting Repetition (ACER). This is a novel compression method that employs dictionary-based techniques to encode repetitive musical sequences that naturally occur within musical audio. As such, it is a lossy compression technique that exploits human perception to achieve data reduction.To evaluate the output from the ACER approach, we conduct a pilot evaluation of the ACER coded audio, by employing both objective and subjective testing, to validate the ACER approach. Results show that the ACER approach is capable of producing compressed audio that varies in subjective and objective and quality grades that are inline with the amount of compression desired; configured by setting a similarity threshold value. Several lessons are learned and suggestion given as to how a larger, enhanced series of listening tests will be taken forward in future, as a direct result of the work presented in this paper.