Algorithmic music generation by harmony recombination with genetic algorithm

Algorithmic music composition has recently become an area of prestigious research in projects such as Google’s Magenta, Aiva, and Sony’s CSL Lab aiming to increase the composers’ tools for creativity. There are advances in systems for music feature extraction and generation of harmonies with short-time and long-time patterns of music style, genre, and motif. However, there are still challenges in the creation of poly-instrumental and polyphonic music, pieces become repetitive and sometimes these systems copy the original files. The main contribution of this paper is related to the improvement of generating new non-plagiary harmonic developments constructed from the symbolic abstraction from MIDI music non-labeled data with controlled selection of rhythmic features based on evolutionary techniques. Particularly, a novel approach for generating new music compositions by replacing existing harmony descriptors in a MIDI file with new harmonic features from another MIDI file selected by a genetic algorithm. This allows combining newly created harmony with a rhythm of another composition guaranteeing the adjustment of a new music piece to a distinctive genre with regularity and consistency. The performance of the proposed approach has been assessed using artificial intelligent computational tests, which assure goodness of the extracted features and shows its quality and competitiveness.

Conventional and Algorithmic Music Listening before Radiotherapy Treatment: A Randomized Controlled Pilot Study

Music listening is a widespread approach in the field of music therapy. In this study, the effects of music listening on anxiety and stress in patients undergoing radiotherapy are investigated. Sixty patients with breast cancer who were candidates for postoperative curative radiotherapy were recruited and randomly assigned to three groups: Melomics-Health (MH) group (music listening algorithmically created, n = 20); individualized music listening (IML) group (playlist of preferred music, n = 20); no music group (n = 20). Music listening was administered for 15 min immediately before simulation and during the first five radiotherapy sessions. The State-Trait Anxiety Inventory (STAI) and the Psychological Distress Inventory (PDI) were administered before/after treatment. Cochran’s Q test and McNemar test for paired proportions were performed to evaluate if the proportion of subjects having an outcome score below the critical value by treatment and over time was different, and if there was a change in that proportion. The MH group improved in STAI and PDI. The IML group worsened in STAI at T1 and improved STAI-Trait at T2. The IML group worsened in PDI at T2. The No music group generally improved in STAI and PDI. Clinical and music listening-related implications are discussed defining possible research perspectives in this field.

Research on the Interaction of Genetic Algorithm in Assisted Composition

Computer-aided composition is an attempt to use a formalized process to minimize human (or composer) involvement in the creation of music using a computer. Exploring the problem of computer-aided composition can enable us to understand and simulate the thinking mode of composers in the special process of music creation, which is an important application of artificial intelligence in the field of art. Feature extraction on the MIDI files has been introduced in this paper. Based on the genetic algorithm in this paper, a platform of the sampling coding method to optimize the character representation has solved the traditional algorithmic music composition study. Music directly from the pitch and duration can be derived from the characteristics, respectively, in the form of a one-hot encoding independently said. Failure to the rhythm of the characterization of the pitch and duration are problems that lead to the inability of compositional networks to learn musical styles better. Rhythm is the combination of pitch and time values according to certain rules. The rhythm of music affects the overall style of music. By associating the pitch and time value coding, the rhythm style of music can be preserved better so that the composition network can learn the style characteristics of music more easily.

Effects of Algorithmic Music on the Cardiovascular Neural Control

Music influences many physiological parameters, including some cardiovascular (CV) control indices. The complexity and heterogeneity of musical stimuli, the integrated response within the brain and the limited availability of quantitative methods for non-invasive assessment of the autonomic function are the main reasons for the scarcity of studies about the impact of music on CV control. This study aims to investigate the effects of listening to algorithmic music on the CV regulation of healthy subjects by means of the spectral analysis of heart period, approximated as the time distance between two consecutive R-wave peaks (RR), and systolic arterial pressure (SAP) variability. We studied 10 healthy volunteers (age 39 ± 6 years, 5 females) both while supine (REST) and during passive orthostatism (TILT). Activating and relaxing algorithmic music tracks were used to produce possible contrasting effects. At baseline, the group featured normal indices of CV sympathovagal modulation both at REST and during TILT. Compared to baseline, at REST, listening to both musical stimuli did not affect time and frequency domain markers of both SAP and RR, except for a significant increase in mean RR. A physiological TILT response was maintained while listening to both musical tracks in terms of time and frequency domain markers, compared to baseline, an increase in mean RR was again observed. In healthy subjects featuring a normal CV neural profile at baseline, algorithmic music reduced the heart rate, a potentially favorable effect. The innovative music approach of this study encourages further research, as in the presence of several diseases, such as ischemic heart disease, hypertension, and heart failure, a standardized musical stimulation could play a therapeutic role.

Algorithmic Music for Therapy: Effectiveness and Perspectives

This study assessed the short-term effects of conventional (i.e., human-composed) and algorithmic music on the relaxation level. It also investigated whether algorithmic compositions are perceived as music and are distinguishable from human-composed music. Three hundred twenty healthy volunteers were recruited and randomly allocated to two groups where they listened to either their preferred music or algorithmic music. Another 179 healthy subjects were allocated to four listening groups that respectively listened to: music composed and performed by a human, music composed by a human and performed by a machine; music composed by a machine and performed by a human, music composed and performed by a machine. In the first experiment, participants underwent one of the two music listening conditions—preferred or algorithmic music—in a comfortable state. In the second one, participants were asked to evaluate, through an online questionnaire, the musical excerpts they listened to. The Visual Analogue Scale was used to evaluate their relaxation levels before and after the music listening experience. Other outcomes were evaluated through the responses to the questionnaire. The relaxation level obtained with the music created by the algorithms is comparable to the one achieved with preferred music. Statistical analysis shows that the relaxation level is not affected by the composer, the performer, or the existence of musical training. On the other hand, the perceived effect is related to the performer. Finally, music composed by an algorithm and performed by a human is not distinguishable from that composed by a human.

Advanced Algorithmic Music Systems

This chapter builds on the previous discussion of procedural music, examining the advanced capabilities of custom-built algorithmic music systems, and how they can respond to minute changes in game data. This exploration examines Uurnog Uurnlimited (2017), a game developed by Niklas Nygren (aka Nifflas), which features such an algorithmic music system. Nifflas designed Uurnog with real-time audio generation as a key feature of the gameplay experience, even going so far as to create a stand-alone music system called Ondskan to give himself nearly unlimited control over how a procedural musical score is attached to game data. Through the exploration of advanced reactive music systems, the reader will encounter many tools useful in thinking outside the box when crafting a music design for a game. These tools can give developers multiple approaches for creating scores that adapt to each play-through and provide the player with a uniquely personalized score.