Music Playlists Research Articles

Surveying More Than Two Decades of Music Information Retrieval Research on Playlists

In this paper, we present an extensive survey of music information retrieval (MIR) research into music playlists. Our survey spans more than 20 years, and includes around 300 papers about playlists, with over 70 supporting sources. It is the first survey that is self-contained in the sense that it combines all the different MIR research into playlists. It embraces topics such as algorithms for automatic generation, for automatic continuation, for assisting with manual generation, for tagging and for captioning. It looks at manually constructed playlists, both those that are constructed for and by individuals, and those constructed in collaboration with others. It covers ground-breaking research into enhancing playlists by cross-fading consecutive songs and by interleaving consecutive songs with speech, similar to what happens on a radio show. Most significantly, it is the first survey that can fully incorporate the paradigm shift that has taken place in the way people consume recorded music: the shift from physical media to music streaming. This has wrought profound changes in the size of music collections available to listeners and thus the algorithms that support the construction, curation and presentation of playlists and the methods adopted by users when they also construct, curate and listen to playlists.

Everyday artificial intelligence unveiled: Societal awareness of technological transformation

Research background: As Artificial Intelligence (AI) weaves into the fabric of daily life, its societal and economic implications underscore the urgency of embracing an environment conducive to its informed adoption. This requires a sophisticated understanding of the societal perception and adaptability to AI, emphasizing the importance of developing comprehensive AI literacy. Purpose of the article: This study inquiries into the sociodemographic underpinnings of AI literacy, aiming to demystify how knowledge about AI's capabilities in everyday tasks varies across individual population segments. It allows us to define the basic determinants that influence the differences in the individual population structures. It also reveals the potential risks associated with the use of AI. Methods: This study investigates the awareness of Artificial Intelligence (AI) in daily lives of the Czech population, focusing on the influence of socio-demographic factors. Utilizing computer-assisted web interviewing, we surveyed 1,041 respondents in April 2023, ensuring representativeness by applying quotas for age, gender, education, region, and residential area size. Our investigation spanned AI applications in sectors like customer service, music playlist recommendation, email sorting, healthcare, online shopping, and home devices. Findings & value added: Findings taken from descriptive statistics reveal variable AI awareness levels across different domains, with younger demographics exhibiting notably lower awareness in several areas. Regression analysis highlighted that awareness is significantly associated with gender, age, and education level. Regression analysis showed that males, younger age groups and those with higher levels of education were more likely to correctly answer majority of questions about the role of AI in everyday life. These insights are crucial for stakeholders aiming to enhance AI literacy, tailor communication strategies, and develop digital platforms, offering guidance for policymakers and market analysts in optimizing AI-related initiatives.

EmoTunes: Your Emotional Soundtrack

The human face is an important organ of an individual’s body and it especially plays an important role in extraction of an individual‘s behavior and emotional state. Manually segregating the list of songs and generating an appropriate playlist based on an individual’s emotional features is a very tedious, time consuming, labor intensive and upheld task. Various algorithms have been proposed and developed for automating the playlist generation process.However the proposed existing algorithms in use are computationally slow, less accurate and sometimes even require use of additional hardware like EEG or sensors. This proposed system based on facial expression extracted will generate a playlist automatically thereby reducing the effort and time involved in rendering the process manually. Thus the proposed system tends to reduce the computational time involved in obtaining the results and the overall cost of the designed system, thereby increasing the overall accuracy of the system. Testing of the system is done on both user dependent (dynamic) and user independent (static) dataset. Facial expressions are captured using an inbuilt camera. The accuracy of the emotion detection algorithm used in the system for real time images is around 85-90%, while for static images it is around 98- 100%.The proposed algorithm on an average calculated estimation takes around 0.95-1.05 sec to generate an emotion based music playlist. Thus, it yields better accuracy in terms of performance and computational time and reduces the designing cost, compared to the algorithms used in the literature survey. Keywords—Audio Emotion Recognition, Music Information Retrieval, Emotion Extraction Module, Audio Feature Extraction Module, Artificial Neural Networks, Confusion Matrix, Viola and Jones Face Detection

Music-making in microgravity: across the first 63 years (1961-2024) of spaceflight. Academy transaction note

Musical expression is a characteristic of humanity on Earth. This paper highlights the historical relationship between music and spaceflight, including its use in the representation of space, and its role in space culture. The paper documents circa 57 musical instruments: from a miniaturized harmonica to a 1.5 m Digeridoo that are known to have been taken into space – usually as part of a crew members personal allowance. Due to mass and volume restrictions numerous classical and ethnographic instruments have been ‘spaceflight-adapted’, in addition to the use of digital technologies for personal musical pleasure and public relations. Furthermore, it charts how access to personalised music playlists and shared musical experiences have been deemed to be vital by crew, promoting individual wellbeing including a connection to home, to each other and playing a part in the marking of success, and tragedy. The value of music making in space is briefly described leading to the proposition that as humanity ventures with greater regularity to Low Earth Orbit and beyond, consideration of how music can be effectively utilised to support individual and crew wellbeing, whilst contributing to the cannon of human artistic expression is warranted.

EMOTION BASED MUSIC RECOMMENDATION SYSTEM

The daily lives of people are greatly influenced by music. Everyone wants to listen to music that suits their personal tastes and mood. Users are constantly needed to manually browse the music and construct a playlist based on their mood. The suggested project, which creates a music playlist depending on users' current moods, is quite effective. The most effective approach to convey a person's current mood is through their facial expressions. Webcams are used to capture facial expressions, which are then fed into a learning algorithm to determine the most likely emotion. The emotion recognition system is built using the FER 2013 dataset for training, which allows it to identify seven different emotions. Its operation involves capturing live video input from a webcam, processing it through the model, and then providing a prediction of the detected emotion. By employing webcams to capture real-time facial expressions, the system dynamically adapts playlist recommendations to match the user's emotional state. This seamless integration of emotion recognition technology with music selection enhances user experience by ensuring that the music aligns with their current feelings, thereby fostering a more immersive and personalized listening experience.

Real Time Emotion Based Music Player

Abstract: In this paper, we present the development and implementation of an Emotion-based Music Player System utilizing facial emotion data, Convolutional Neural Networks (CNN), Flask, OpenCV for face detection, and Spotify API for music playlist integration. The system aims to provide users with personalized music recommendations based on their current emotional state, detected through real-time analysis of facial expressions. The proposed system consists of several key components, including data collection, preprocessing, CNN algorithm implementation for emotion classification, integration with Flask for web application development, and real-time emotion detection using OpenCV. The trained CNN model is capable of accurately classifying emotions such as anger, disgust, fear, happiness, neutrality, sadness, and surprise. Results from the implementation demonstrate the system's effectiveness in accurately detecting emotions from facial expressions and providing corresponding music recommendations. The CNN model achieved a final accuracy of 62.44% after 10 epochs of training. Realtime emotion detection using OpenCV successfully identifies facial expressions, allowing for dynamic adjustments to the music playlist. Overall, the Emotion-based Music Player System presents a novel approach to enhancing user experience by leveraging facial emotion data and advanced machine learning techniques to deliver personalized music recommendations tailored to individual emotional states.

Music as a countermeasure to fatigue: A driving simulator study

Listening to music while driving is a commonly adopted fatigue-coping strategy, yet its effectiveness has been explored in only a limited number of studies. Here, we aim to evaluate its impact on passive task-related driving fatigue, as measured by a set of both objective and subjective indicators.A driving simulator experiment was conducted involving 60 participants, who completed a 50-minute driving task on a monotonous highway scenario. The participants were divided into a control group, which did not listen to any music, and an experimental group, which drove for 20 min without music and then listened to a self-selected music playlist played at the preferred intensity for the remaining 30 min.Results show that music had a positive acute effect on objective measurements of driving fatigue. However, this effect was transient (15–25 min duration), and more prominently observed in subjective measures of fatigue. An exploratory analysis of song features suggests that higher intensity, higher tempo, more danceable, and less instrumental music tend to be more effective. These findings indicate that listening to music can serve as a viable short-term countermeasure to driving fatigue.

Implementation Music Recommendation Based on Facial Expression

We proposed a new approach for automatically playing music using facial emotions. Most existing approaches involve manually playing music, using wearable computing devices, or classifying them based on audio features. We used a Convolutional Neural Network foremotion detection. Pygame and Tkinter were used for music recommendations. Facial expressions were captured using a built-in camera. Feature extraction is performed on input face images to detect emotions such as happy, angry, sad, surprised, and neutral. An automatic music playlist was generated by identifying the current emotions of the user.

Music Recommendation Based On Facial Expression

We proposed a new approach for automatically playing music using facial emotions. Most existing approaches involve manually playing music, using wearable computing devices, or classifying them based on audio features. We used a Convolutional Neural Network foremotion detection. Pygame and Tkinter were used for music recommendations. Facial expressions were captured using a built-in camera. Feature extraction is performed on input face images to detect emotions such as happy, angry, sad, surprised, and neutral. An automatic music playlist was generated by identifying the current emotions of the user.

CAN YOU FIND THE MARKETING IN THE MUSIC? AN INTERACTIVE CONCEPT-MATCHING ACTIVITY

ABSTRACT Student engagement is a growing challenge within higher education, especially as technology continues to evolve and competition for attention rises. As a business discipline, many marketing students struggle to connect with the theory presented in the classroom and seek interactive, experiential learning opportunities. The authors created the Find the Marketing in the Music playlist matching activity to address this challenge and provide students with an engaging application of marketing concepts. Music is a means of personal expression and broadens listeners’ perspectives. Music can help ignite or explain feelings, highlight complexities of social situations, and serve as a vehicle driving understanding and compassion. This featured activity reveals that by pairing songs that illustrate marketing concepts with information discussed in class, students could connect to the material in a new way. This exercise allows students to leverage the creative and grounded aspects of critical thinking by connecting something artistic with concrete marketing concepts. Students note that this activity helps them better relate to the material and that the relevancy of things discussed in the classroom takes on a new level when integrated into their everyday lives.

The effect of music type in ketamine-assisted group therapy on treatment-resistant mental health conditions: a prospective observational study

Introduction Currently, Ketamine is the only safe, effective, and widely used psychedelic-like medicine in Canada. It has demonstrated notably efficacy in providing relief to those experiencing treatment resistant mental health conditions. Pairing Ketamine treatment with psychotherapy, known as Ketamine Assisted Therapy (KAT), has been shown to yield more enduring outcomes. Work by Greenway et al. has demonstrated that playing music following ketamine administration for patients with bipolar disorder can help the patient feel more in control and reduce discomfort (Greenway et al. International Clinical Psychopharmacology 2021; 36 218-220).ObjectivesThe primary objective is to evaluate and compare the subjective clinical efficacy of two different types of music during ketamine-assisted group therapy. This will be explored through various validated psychiatric questionnaires, including the PHQ-9, GAD-7, and PCL-5. The secondary objective is to compare the objective changes in brain activity between the two music types. This will be evaluated using EEG data collected from MUSE headband before and after each ketamine-assisted therapy session.Methods This study is a crossover trial of 32 participants undergoing ketamine-assisted therapy for treatment-resistant mental health conditions. Half of participants will undergo a KAT session with a “weightless” music playlist followed by a session with a “grounding” music playlist. The other half will do the same, in reverse order. All participants will complete several psychiatric questionnaires within 7 days of each session over email. Before and after each session, participants will play a simple game to test executive function while wearing a headband to measure EEG activity.ResultsThe absolute and relative changes to the scores of the questionnaires will be examined between participants and music conditions. The change in brain activity from pre-session to post-session will be compared between the different music conditions as well. As this is a crossover trial, any changes in outcomes due to order effects will be controlled for. Relevant demographic and medical factors will also be controlled for.ConclusionsTo date, no studies have explored the influence that different types of music have on patients experience with KAT in a group therapy setting. With the results of this study, we hope to fine tune and improve the use of music in future KAT administration.Disclosure of InterestNone Declared

Music Recommendation System based on Facial Expressions by using CNN

Abstract: In today's world, the integration of emotion detection and personalized music recommendation has become increasingly prevalent across diverse fields. This project endeavors to develop a cutting-edge music recommendation system that tailors playlists to users' emotional states, as discerned from their facial expressions in real-time. Leveraging the Facial Expression Recognition 2013 (FER 2013) dataset, which encompasses a wide array of emotional expressions including happiness, sadness, anger, fear, disgust, neutrality, and surprise, serves as the foundation for training our model. The project employs the Mediapipe algorithm for feature extraction from facial expressions and utilizes a Convolutional Neural Network (CNN) to classify these emotions accurately. For seamless integration and user interaction, streamlit web browser frameworks are harnessed to facilitate music recommendations within an intuitive interface. Furthermore, the proposed system capitalizes on real-time face detection through camera input using OpenCV, ensuring timely and responsive interaction. By automatically generating music playlists based on the user's current emotional state, our system aims to significantly reduce computational time and overall costs compared to existing literature. In essence, this project presents a comprehensive solution that combines real-time facial expression analysis with personalized music recommendation, offering enhanced efficiency and accuracy while minimizing computational overhead.

Study protocol: randomized controlled trial of an individualized music intervention for people with dementia in the home care setting

BackgroundStudies suggest that individualized music listening is an effective, non-pharmacological intervention for improving the quality of life of people with dementia in the institutional care setting. Noting that most people with dementia live at home, we conduct a randomized controlled trial to assess the feasibility and effectiveness of an app-based individualized music listening intervention for people with dementia in the home care setting. The intervention is delivered by family caregivers.MethodsWe will recruit N = 130 dyads consisting of one person with dementia living at home and their family caregiver. After a baseline assessment, dyads are randomly assigned by gender to either the intervention or control group. People with dementia in the intervention group listen to individualized music playlists for 20 min every other day for six weeks via the self-developed Individualized Music and Dementia app. The control group receives standard care. All dyads complete paper-and-pencil questionnaires six weeks before the start of the intervention (T0), directly before the intervention (T1), directly after the intervention (T2), and six weeks later (T3). During the intervention period, all caregivers also complete daily ecological momentary assessments via the app. During three home visits, a trained project member will observe the dyads and collect hair samples. After the intervention, semi-structured interviews will be conducted to collect information about participants’ experiences with the app and intervention. The primary outcome is the attainment of individual goals established during the baseline assessment. Secondary outcomes are the well-being, physiological stress and quality of life of people with dementia and their caregivers; people with dementia’s behavioural and psychological symptoms of dementia, resistance during care, and reactions to the music; caregivers’ burden of care, positive aspects of care, and caregiving self-efficacy; and the quality of the caregiver-care recipient interaction.DiscussionOur study will assess the extent to which an app-based individualized music listening intervention is feasible and effective for enhancing the well-being and quality of life of people with dementia living at home and their family caregivers.Trial registrationGerman Clinical Trials Register DRKS00025502 and ISRCTN registry ISRCTN68084105, https://doi.org/10.1186/ISRCTN68084105

Emotion equalization app: The effectiveness of dynamic music therapy approaches

A recent study explored how static music playlists of relaxing or uplifting music can improve our mood. Unlike static playlists that rely exclusively on relaxing, uplifting, or consoling music, this study explores eight distinct methods to curate playlists. These approaches include transitions from one dimension to another (e.g., Angry-to-Relaxing, Angry-Uplifting, Relaxing-Uplifting, Relaxing- Natural, Sad-Relaxing, Sad-Uplifting, Uplifting-Relaxing, and Uplifting-Natural). To evaluate the effectiveness of these playlists, online experiments were conducted, wherein participants were surveyed before and after music listening to monitor changes in valence and arousal. The results reveal that Sad-Relaxing, Relaxing-Uplifting, Relaxing-Natural, and Uplifting-Natural playlists give positive changes in participants' valence levels, indicating mood improvement. As for the arousal level, the Angry-Relaxing playlist made people feel calmer. Correspondingly, the Relaxing-Uplifting playlist made people feel more energetic. Moreover, this study has identified specific pathways on a two-dimensional plane where both arousal and valence values could be profitably to good effect, including transitioning from Angry to Relaxing to Uplifting especially for arousal, and Sad to Relaxing or Uplifting especially for valence. These findings highlight how dynamic music can give rise to distinct emotional responses and influence our energy levels, and give guidelines on how to use music therapy playlists for mood enhancement.

2035 The Feasibility to Staff of Patient Specific Music Choice on Elderly Care Wards - A Post Intervention Survey

Abstract Introduction Receptive music listening has been shown to reduce depression, anxiety, and agitation in older adults. However, unfiltered and disruptive noise can increase confusion and agitation. Yet, during hospitalization older patients often have little control over when and for how long they are exposed to music, the genre which is heard, or they may have no access to music. Furthermore, older persons have reduced ability to use modern technology to counter this and their sensory and functional impairments may further isolate them. This study aimed to assess the feasibility of offering two one-hour daily sessions of patient specific music (PSM) choices on an elderly-care ward. Methods On a district general hospital elderly-care ward a 5-day trial of offering two one-hour daily sessions of PSM, using enhanced wireless speakers optimally positioned with daily amended music playlists based on specific patient choice. Ambient noise was minimised with regular decibel monitoring. A post-intervention staff survey was completed to assess the feasibility of continuing, the perceived impact on staff and patients and potential barriers to continuation. Thematic analyses were completed on the survey. Results In the post-intervention feasibility survey (n=14) the majority of staff (86%) agreed that it was possible, implementable and the procedure easy to use. In the impact assessment (n=19) 80% of staff stated it had a positive effect on patients and 89% stated it had a positive effect on staff. Thematic analyses on impact identified benefits to: work, engagement, enjoyment, physical activity and well-being. Barriers that were identified included patient choice, repetition of music, patients unable to engage with the process and staff availability for consistent delivery. Conclusion Playing patient specific music choice was feasible and acceptable to staff with a perceived positive influence on both staff and patients. Further studies are now required to assess the impact on patient outcomes.

From listening to curating: Anthropological curatorship toward music playlist practices

Currently, the curatorial approach is expanding its influence beyond the limitations of museums and art galleries, transforming the experiences and significance of our daily existence. That is embodied, for example, in the curatorial practices of music playlists. In making a music playlist there must be a sound curation process where the songs to listen to are selected and included in the music playlist. Accordingly, this study aims to describe how users of analog (mixtapes) and digital (Spotify playlist) music playlists conduct their music curation process and perceive the relationships behind them. The curation process is analyzed using curatorial and anthropological frameworks, which also attempt to provide a fascinating background on preference formation and the curation process. Qualitative data was collected from a virtual ethnographic approach, with observations and unstructured-semi-structured interviews—offline and online—of 19 informants aged 21–55 living in Yogyakarta and Bandung and a literature review. The data obtained through this study focuses on two outcomes. First, sociocultural circumstances provide the basis for preference formation in selecting, curating, and making music playlists. Second, the particulars of each user's music playlist are related to self-discovery and individual identity. As a result, sociocultural circumstances persuade music playlist users' habits that attach to the curation process.

Online music listening programs for older adults with dementia during the COVID-19 Pandemic: A feasibility study

Older adults and people living with dementia experienced high degrees of social isolation and reduced access to support during the recent COVID-19 pandemic. However, worldwide, including amongst people living with dementia and their caregivers, musical engagement provided some relief and social connection. This study examines the feasibility of transforming a personalised playlist program for people living with dementia for online delivery. A mixed methods approach was taken in which quantitative data (facial action unit activation) and qualitative data (interviews) were collected from 5 people with dementia and their caregivers. Interviews were also conducted with 5 music therapy students undergoing clinical placements who participated in the study as program facilitators. Results demonstrated the feasibility of delivery of online music playlist programs for supporting people living with dementia living in isolation. Caregivers reported increased understanding about how to use music in targeted ways to support mood and behaviour. Despite the limitations of the online format, facial action unit activation suggested that people with dementia experienced increased enjoyment when listening to energising playlists, confirming the feasibility of this as an outcome measure. However, increased challenges to developing a rapport with clients in the online context were highlighted.

A subjective evaluation of different types of music in music therapy using an emotion equalization app

Music therapy has been shown to be effective at improving mood (i.e., valence) in subjects in previous studies. In this study, we investigate five different ways of generating playlists of therapeutic music to see which is most effective at improving mood, tested over a large number of subjects. We conducted online experiments under three proposed approaches using playlists of consoling, relaxing, and uplifting music, and compared them to two baseline methods using natural sounds and a random mix of songs (i.e., music that could be either relaxing, uplifting, or consoling). We questioned subjects before and after listening to the music to see how valence and arousal changed, and analyzed the results with ANOVA. The results showed there were positive changes in listeners’ valence levels for natural sounds, as well as for consoling, relaxing, and uplifting music playlists. There were no significant changes in arousal for most sources, but there was a lowering in arousal with relaxing music and a rising in arousal with uplifting music. Additionally, relaxing and uplifting music playlists both showed significant effects in moving listeners from negative to more positive emotional states.

Individualized Music Interventions for Dementia-Related Behavior Management.

Dementia presents many challenges for the person diagnosed, as well as for caregivers. The use of psychotropic medications for dementia-related behaviors has been linked to adverse reactions and increased mortality. Alternative approaches such as the use of personalized music playlists and musically-related spiritual care have had promising results including improved quality of life for persons with dementia and reduced stress for caregivers. Implementing an individualized music program is relatively simple and avoids deleterious side effects that may accompany medication use.

Nature-themed video intervention may improve cardiovascular safety of psilocybin-assisted therapy for alcohol use disorder.

Psychedelic-assisted therapy with psilocybin has shown promise in Phase 2 trials for alcohol use disorder (AUD). Set and setting, particularly factors facilitating a connection with nature, may positively influence the psychedelic experience and therapeutic outcomes. But to date, randomized controlled trials of interventions to enhance set and setting for psychedelic-assisted therapy are lacking. This was a pilot randomized, controlled trial of Visual Healing, a nature-themed video intervention to optimize set and setting, versus Standard set and setting procedures with two open-label psilocybin 25 mg dosing sessions among 20 participants with AUD. For the first session, participants randomized to Visual Healing viewed nature-themed videos during the preparation session and the "ascent" and "descent" phases of the psilocybin dosing session while participants randomized to the Standard condition completed a meditation during the preparatory session and wore eyeshades and listened to a music playlist throughout the dosing session. For the second session 4 weeks later, participants chose either Visual Healing or Standard procedures. Primary outcomes were feasibility, safety, and tolerability of Visual Healing. Secondary and exploratory outcomes were changes in alcohol use, psychedelic effects, anxiety and stress. Nineteen of 20 (95%) randomized participants (mean age 49 ± 11 years, 60% female) completed the 14-week study. During the first psilocybin session, participants viewed an average of 37.9 min of the 42-min video and there were no video-related adverse events. Peak increase in post-psilocybin blood pressure was significantly less for participants randomly assigned to Visual Healing compared to Standard procedures. Alcohol use decreased significantly in both Visual Healing and Standard groups and psychedelic effects, stress, and anxiety were similar between groups. In this open-label pilot study, viewing Visual Healing videos during preparation and psilocybin dosing sessions was feasible, safe, and well-tolerated among participants with AUD. Preliminary findings suggest that Visual Healing has potential to reduce the cardiovascular risks of psychedelic therapy, without interfering with the psychedelic experience or alcohol-related treatment outcomes. Studies to replicate our findings as well as studies of different set and setting interventions with other psychedelic medications and indications are warranted.