Sistem Klasifikasi Musik Gamelan Angklung Bali Terhadap Suasana Hati Menggunakan Algoritma K-Nearest Neighbor Berbasis Algoritma Genetika

Musik instrumen gamelan angklung Bali lewat gelombang bunyi yang dihasilkannya mampu menginterferensi gelombang pikiran manusia untuk menurunkan frekuensi gelombang yang dipancarkan oleh otak. Tujuannya untuk mempengaruhi kondisi psikologi yang berkaitan dengan suasana hati agar mengarah pada tingkat stress positif dengan tingkat energi rendah maupun tinggi. Musik dengan tingkat stress positif dan tingkat energi rendah masuk ke dalam kategori suasana hati tenang atau contentment, jika tingkat stress positif dan tingkat energi tinggi masuk ke dalam kategori suasana hati senang atau exuberance. MIR (Music Information Retrieval) adalah bagian dari Data Mining yang menggali informasi mengenai data musik, salah satunya yaitu klasifikasi suasana hati yang diinterpretasikan oleh potongan data musik. Penelitian ini merancang dan membangun sistem klasifikasi untuk mendeteksi suasana hati musik instrumen gamelan angklung Bali menggunakan algoritma K-NN dan K-NN berbasis Algoritma Genetika. K-NN dapat mengatasi masalah klasifikasi dengan baik, namun dibalik keunggulannya, pengaturan nilai k yang sangat sensitif menjadi sebuah kelemahan. Menerapkan operasi genetika oleh Algoritma Genetika pada sistem klasifikasi K-NN berhasil mengoptimasi penentuan nilai k optimal, serta memperbaiki hasil akurasi klasifikasi. Berdasarkan dataset training dan dataset testing yang sama, K-NN memberikan persentase akurasi tertinggi sebesar 81,08% (k=6), sedangkan K-NN berbasis Algoritma Genetika memberikan persentase akurasi tertinggi sebesar 89,19% (k=4).

Enabling FAIR use of Ethnomusicology Data – Through Distributed Repositories, Linked Data and Music Information Retrieval

Recordings of musical practices are kept in various public institutions and private depositories around the world. They constitute valuable data for ethnomusicological research and are substantial for the world's musical heritage. At the moment, there are no commonly used systems and standards for organizing, describing or categorizing these data, which makes their use difficult. In this paper, we discuss the required steps to make them findable, accessible, interoperable and reusable (FAIR), and outline action items to reach these goals. We show solutions that help researchers to manage their data over the whole research lifecycle and discuss the benefits of combining technologies from information science, music information retrieval, and linked data, with the aim of giving incentives for the ethnomusicology research community to actively participate in these developments in the future.

Audio classification for music information retrieval of Hindustani vocal music

An important task in music information retrieval of Indian art music is the recognition of the larger musicological frameworks, called ragas, on which the performances are based. Ragas are characterized by prominent musical notes, motifs, general sequences of notes used and embellishments improvised by the performers. In this work we propose a convolutional neural network-based model to work on the mel-spectrograms for classication of steady note regions and note transition regions in vocal melodies which can be used for finding prominent musical notes. It is demonstrated that, good classification accuracy is obtained using the proposed model.

MUHSIC: An Open Dataset with Temporal Musical Success Information

Music is an alive industry with an increasing volume of complex data that creates new challenges and opportunities for extracting knowledge, benefiting not only the different music segments but also the Music Information Retrieval (MIR) community. In this paper, we present MUHSIC, a novel dataset with enhanced information on musical success. We focus on artists and genres by combining chart-related data with acoustic metadata to describe the temporal evolution of musical careers. The enriched and curated data allow building success-based time series to investigate high-impact periods (hot streaks) in such careers, transforming complex data into knowledge. Overall, MUHSIC is a relevant tool in music-related tasks due to its easy use and replicability.