Abstract:Abstract: The rapid growth of the digital music industry requires accurate music genre classification systems to enhance user experience in streaming services. This study compares a domain-specific Long Short-Term Memory…
(LSTM) network with three Large Language Models (LLMs)—HuBERT, WavLM, and WAV2Vec 2.0—for Music Genre Classification (MGC). The LSTM model was trained using Mel-spectrograms transformed from the GTZAN dataset, while the LLMs were fine-tuned using a smaller set of raw audio samples due to computational constraints. All models were tested on datasets with identical genre labels to ensure a fair evaluation. Results show that the LSTM model achieved the highest accuracy of 97.10%, outperforming HuBERT (86.00%), WavLM (83.00%), and WAV2Vec 2.0 (80.00%). The LSTM demonstrated superior generalization and stability without overfitting, while the LLMs struggled to differentiate between genres with similar acoustic characteristics. These findings indicate that general-purpose pre-trained models, although powerful, are less effective in music-specific tasks due to domain mismatch. Therefore, incorporating music-specific features and architectures remains essential for achieving higher accuracy and reliability in automatic genre classification systems.
Keywords: audio large language models; comparative deep learning; music genre classification.
Abstrak: Pertumbuhan industri musik digital yang pesat menuntut sistem klasifikasi genre musik yang akurat untuk meningkatkan pengalaman pengguna dalam layanan streaming. Penelitian ini dilatarbelakangi oleh perkembangan pesat model pembelajaran mendalam, khususnya jaringan LSTM dan model bahasa berskala besar LLM seperti HuBERT, WavLM, dan WAV2Vec 2.0, yang telah menunjukkan kemampuan representasi audio yang kuat. Tujuan penelitian ini ini membandingkan jaringan Long Short-Term Memory (LSTM) khusus domain dengan tiga model Large Language Models (LLM)—HuBERT, WavLM, dan WAV2Vec 2.0—untuk tugas Klasifikasi Genre Musik (MGC). Metode penelitian melibatkan pelatihan LSTM menggunakan data Mel-spectrogram hasil transformasi dari dataset GTZAN, sementara LLM disesuaikan (fine-tuning) menggunakan data audio mentah dalam jumlah lebih kecil karena keterbatasan komputasi. Seluruh model diuji pada dataset dengan label genre yang sama untuk memastikan evaluasi yang adil. Hasil penelitian menunjukkan bahwa model LSTM mencapai akurasi tertinggi sebesar 97,10%, sedangkan model HuBERT, WavLM, dan WAV2Vec 2.0 masing-masing memperoleh 86,00%, 83,00%, dan 80,00%. Model LSTM menunjukkan kemampuan generalisasi yang lebih baik tanpa overfitting, sedangkan model LLM cenderung kesulitan membedakan genre dengan karakteristik akustik yang mirip. Kesimpulan penelitian ini adalah ketidaksesuaian domain secara signifikan membatasi performa model umum saat diterapkan pada tugas berbasis musik. Oleh karena itu, penggunaan fitur dan arsitektur khusus musik sangat penting dalam membangun sistem klasifikasi genre yang lebih akurat.
Kata kunci: klasifikasi genre musik; model bahasa besar; perbandingan pembelajaran mendalam.
Abstract:Abstract: Along with the development of information technology, the way people listen to music also changed with the emergence of many music streaming applications. The user experience of an application plays an important…
t role in attracting and retaining application users. Therefore, this study has goal to compare the music steaming applications’ user experience of the Joox and Spotify by using User Experience Questionnaire (UEQ) method. There are 6 scales assessed by UEQ, including attractiveness, perspicuity, efficiency, dependability, stimulation, and novelty. Data was collected by distributing online questionnaires and obtained as many as 104 responses that could be processed. The results show that both applications have positive impressions which are indicated by the average value of each scale greater than 0.8. This indicates that these two applications are considered quite fun, effective, easy to understand, and innovative. However, the comparison’ results show that between Spotify and Joox is not significantly different, with Spotify scores slightly better than Joox.
Keywords: music streaming application; user experience; user experience questionnaire
Abstrak: Seiring dengan perkembangan teknologi informasi, cara masyarakat mendengarkan musik juga mengalami perubahan dengan munculnya banyak aplikasi streaming musik. User experience suatu aplikasi berperan penting untuk menarik dan mempertahankan pengguna aplikasi. Oleh karena itu, penelitian ini memiliki tujuan untuk membandingkan user experience dari aplikasi streaming musik Joox dan Spotify dengan metode User Experience Questionnaire (UEQ). Terdapat 6 skala yang dinilai oleh UEQ yaitu daya tarik, kejelasan, efisiensi, ketepatan, stimulasi, dan kebaruan. Data dikumpulkan melalui penyebaran kuesioner secara daring dan diperoleh sebanyak 104 jawaban yang bisa diolah. Hasil penelitian menunjukkan bahwa kedua aplikasi memiliki impresi positif yang ditandai dengan nilai rata-rata setiap skala lebih besar dari 0.8. Artinya kedua aplikasi ini dinilai cukup menyenangkan, efektif, mudah dipahami, dan inovatif. Meskipun demikian, hasil perbandingan menunjukkan bahwa perbedaan antara Spotify dan Joox tidak terlalu signifikan, dengan nilai Spotify sedikit lebih unggul daripada Joox.
Kata kunci: aplikasi streaming musik; pengalaman pengguna; user experience questionnaire
Abstract:This study explores the cultural practice of fangirling on social media platform X as a space for identity negotiation and communal solidarity in the digital era. Using a digital ethnography approach, data was collected…
through participant observation, in-depth interviews with fans, and a documentary study of the digital footprints of the informants. The results show that fangirling is not merely a form of passive consumption, but rather a communal cultural activity carried out massively and organized through fictitious kinship structures in cyberspace. This phenomenon manifests itself in the form of digital "work culture," such as the ritual of collectively streaming music and mobilizing votes to increase the symbolic capital of the idol group SEVENTEEN on the global stage. This practice confirms a shift in social interaction patterns, where group loyalty is bound by a commitment to the idol's aesthetics and achievements. Anthropologically, this activity creates a new value system that blurs the boundaries between the private space of fans and the digital public space, while simultaneously strengthening the position of social media platform X as a primary locus in the formation of contemporary popular culture.
Abstract:Pariwisata telah menjadi fokus utama pembangunan di Indonesia sejak lama, sebagaimana tercermin dalam instruksi Presiden Republik Indonesia Nomor 9 tahun 1969. Pengembangan pariwisata dianggap sebagai bagian integral dari…
i upaya pengembangan, pembangunan, dan kesejahteraan masyarakat serta negara. Pemanfaatan social media Instagram bisa dibilang sarana yang cukup efisien dalam mempromosikan pariwisata. Oleh karena itu dalam memperkenalkan wisata puncak rembangan terhadap pengunjung yang masih awam melalui postingan Instagram. Tujuan dari penelitian ini ialah untuk menerapkan pemanfaatan media social Instagram dalam memperkenalkan wisata puncak rembangan terhadap wisatawan yang ada dikota jember. Penelitian ini menggunakan metode kualitatif dengan teknik observasi partisipan untuk mengumpulkan data terkait pemanfaatan perkembangan digitalisasi dalam promosi pariwisata, khususnya di destinasi wisata Puncak Rembangan. Hasil dari penelitian ini menunjukkan bahwa promosi melalui media sosial khususnya isntagram telah berhasil meningkatkan minat wisatawan untuk berkunjung ke Puncak Rembangan. Adanya promosi yang diadakan, serta acara live music pada event tertentu, telah menarik perhatian pengunjung dan meningkatkan jumlah kunjungan wisatawan dari waktu ke waktu.
Abstract:The Independent curriculum is a new curriculum used in Indonesia and a transitional curriculum from K-13 curriculum. Therefore, this study aims to examine how the role of a teacher in implementing the new curriculum in the…
he class they teach, especially English teachers. This study used library research methods, the data were taken from several articles and similar research. The source of data came from Google Scholar by using keywords the role of the teacher in the independent curriculum and the application of the independent curriculum in English classes. The results of this study indicate that English teachers are people who play an important role in implementing the independent curriculum. The English teacher in this curriculum acts as a giver of understanding, and has influence to improve the quality of education as well as an adapter that aligns the curriculum with the characteristics and needs of students. English teachers can implement an independent curriculum by changing teaching methods, using new strategies to improve learning outcomes. For curriculum implementation there are several strategies that English teachers can use such as games, music, using interactive learning media when teaching, then you can also use Role-play, Information-gap activities, Jigsaw activities, open-ended discussions and debates.
Abstract:Latar Belakang : Lanjut usia merupakan seseorang yang usianya mencapai lebih dari sama dengan 60 tahun berdasarkan Undang-Undang Nomor 13 Tahun 1998 tentang Kesejahteraan Lanjut Usia (Kemenkes, 2016). Salah satu permasalahan …
ermasalahan kesehatan terkait pada lansia yaitu dengan gangguan tidur atau yang bisa disebut insomnia. Salah satu upaya yang dilakukan untuk mengatasi gangguan tidur (insomnia) menurut penelitian yang dilakukan Walanda, 2023 adalah dengan pemberian terapi musik. Tujuan : Lansia dapat berespon terhadap stimulus yang di berikan oleh perawat yaitu music klasik.Metode Kegiatan : Kegiatan ini dilakukan di Panti Werdha Marfati wisma belakang dengan mendengarkan musik klasik menggunakan speaker. Hasil : Para lansia sangat merespon dengan baik dalam mendengarkan musik klasik dengan cara bernyanyi dan berjoget.
Abstract:This research aims to describe and analyze the Palimanan Style Rumyang Mask Dance using the Theory of Liminality based on the thoughts of Victor Turner. This study uses the Qualitative Content Analysis method because the…
results of this study are in the form of described sentences. The object of this research is the Palimanan Style Rumyang Mask Dance. The cornerstone of the theory used in this study is Victor Turner's Theory of Liminality in analyzing the Palimanan Style Cirebon Rumyang Mask Dance. Data collection using interviews, observations, document studies and library studies. Data collection uses data triangulation techniques. The results obtained in this study are data descriptions, namely data on the presentation form of The Cirebon Palimanan Style Rumyang Mask Dance including: motion, makeup, fashion, accompaniment music, property and floor design. Palimanan Style Rumyang Mask Dance Data is interpreted based on Victor Turner's theory of Liminality with 3 cycles, namely the Pre liminal stage when the dancer has not danced the rumyang mask, the Liminal stage when the dancer pulls the Mask rumyang Cirebon Palimanan Style, and the last stage is Post Liminal after the dancer pulls the Mask Rumyang Cirebon Palimanan Style. Based on the results and discussions, it can be concluded in the Mask Dance of Rumyang Cirebon Palimanan Style There are 3 cycles of Liminality Theory. and the last stage is Post Liminal after the dancer pulls the Mask Rumyang Cirebon Palimanan Style. Based on the results and discussions, it can be concluded in the Mask Dance of Rumyang Cirebon Palimanan Style There are 3 cycles of Liminality Theory. and the last stage is Post Liminal after the dancer pulls the Mask Rumyang Cirebon Palimanan Style. Based on the results and discussions, it can be concluded in the Mask Dance of Rumyang Cirebon Palimanan Style There are 3 cycles of Liminality Theory.
Abstract:Perkembangan teknologi kecerdasan buatan atau Artificial Intelligence (AI) telah memberikan dampak transformasi yang masif di berbagai sektor, termasuk pendidikan. Guru dituntut untuk beradaptasi dengan alat digital guna…
menciptakan suasana belajar interaktif bagi siswa generasi Z dan Alpha. Pengabdian kepada Masyarakat (PkM) ini bertujuan memberikan solusi atas keterbatasan literasi digital guru di SD Brainfor Islamic School melalui pelatihan intensif pemanfaatan Generative AI. Fokus kegiatan meliputi penggunaan AI Teks (ChatGPT, Notion AI) untuk administrasi, serta AI Multimodal (Gemini AI, Canva AI, Suno AI) untuk pembuatan RPP, ilustrasi kelas, dan musik edukatif. Metode pelaksanaan menggunakan pendekatan Participatory Action Learning System (PALS) yang terdiri dari tahap sosialisasi, pelatihan hands-on, dan evaluasi. Hasil kegiatan menunjukkan lonjakan rata-rata pemahaman peserta dari 35% pada tahap pre-test menjadi 88% pada tahap post-test. Seluruh peserta berhasil menghemat 50% waktu penyusunan administrasi dan menghasilkan minimal satu produk media pembelajaran interaktif secara mandiri. Pemanfaatan AI terbukti mereduksi kebosanan belajar siswa dan sangat membantu guru berinovasi.
The development of Artificial Intelligence (AI) technology has brought massive transformation in various sectors, including education. Teachers are required to adapt to digital tools to create interactive learning environments for Generation Z and Alpha students. This Community Service (PkM) aims to provide a solution to the limited digital literacy of teachers at SD Brainfor Islamic School through intensive training on Generative AI. The focus includes the use of Text AI (ChatGPT, Notion AI) for administration, and Multimodal AI (Gemini AI, Canva AI, Suno AI) for creating lesson plans, class illustrations, and educational music. The implementation method uses the Participatory Action Learning System (PALS) approach consisting of socialization, hands-on training, and evaluation stages. The results showed a surge in participants' average understanding from 35% in the pre-test to 88% in the post-test. All participants successfully saved 50% of their administration time and independently produced at least one interactive learning media product. The use of AI is proven to reduce student boredom and greatly helps teachers innovate.
Abstract:Information is the result of data processing in a form that is more useful and more meaningful for the recipient that describes real events (facts) that are used for decision making. The music information system website…
aims to provide various communication and information services about music. The information system used by the music community so far is considered slow and complicated. Both musicians and producers have difficulty communicating directly and the number of individuals who can communicate directly is relatively small. Therefore, the current system cannot meet the demands of the information needs, it needs to be developed so that it can display information and communication facilities about music directly and quickly. In designing the music information system website using Php & MySql Script. Where in the design of this information system itself uses a Database where the use of the Database itself is as a media that manages all data and a place to store the data itself later. The information system itself is designed so that musicians and producers can create their own profiles. This information system also provides communication services about music. The final result achieved from this music information system is as a communication service for the music community, Sharing Content, publication, and promotion from musicians to the public and producers.