Read the full-text of this article for free

Title: Optimisation of intelligent composition technology based on deep neural networks in the context of cultural tourism IoT

Authors: Jiren Wang; Meiqi Zhu; Dayang Chen

Addresses: Music Academy, Beihua University, Jilin City, Jilin Province, 132000, China ' Music Academy, Beihua University, Jilin City, Jilin Province, 132000, China ' Music Academy, Beihua University, Jilin City, Jilin Province, 132000, China

Abstract: This work focuses on optimising intelligent composition technology in the context of the cultural tourism internet of things (IoT); it takes the optimisation of the music generation algorithm as the core goal and improves the quality and innovation of music generation. Utilising Python's Mido library, the main melody of musical instrument digital interface (MIDI) music files is extracted. The skip-gram model of Word2vec is then used to convert the note sequences into feature vectors. The performance of various models is compared, including long short-term memory (LSTM), gated recurrent unit (GRU), bi-directional long short-term memory (Bi-LSTM), bidirectional gated recurrent unit (Bi-GRU), Bi-LSTM + attention, and Bi-GRU + attention. The metrics used for comparison include note accuracy, bilingual evaluation understudy (BLEU) score, creativity, melodic beauty, rhythmic smoothness, and auditory experience, aiming to demonstrate the effectiveness of the proposed model.

Keywords: internet of things; IoT; deep neural network; DNN; intelligent composition; Bi-LSTM; note accuracy; cultural tourism.

DOI: 10.1504/IJART.2026.155624

International Journal of Arts and Technology, 2026 Vol.16 No.9, pp.92 - 111

Received: 15 Mar 2026
Accepted: 10 May 2026
Published online: 07 Aug 2026 *