A5033634140- Music Technology and Sound Studies
- Music and Audio Processing
- Image Processing Techniques and Applications
- Advanced SAR Imaging Techniques
- Diverse Musicological Studies
- Brain Tumor Detection and Classification
- Video Analysis and Summarization
- Handwritten Text Recognition Techniques
- Human Motion and Animation
- Image Processing and 3D Reconstruction
- Human Pose and Action Recognition
- CCD and CMOS Imaging Sensors
Guangdong University of Technology
2025
Beijing University of Posts and Telecommunications
2022-2023
This paper is part of a special issue on Music Technology. We study the type recognition traditional Chinese musical instrument audio in common way. Using MEL spectrum characteristics as input, we train an 8-layer convolutional neural network, and finally achieve 99.3% accuracy. After that, this mainly studies performance skill instruments. Firstly, for single instrument, features were extracted by using pre-trained ResNet model, then SVM algorithm was used to classify all instruments with...
The development of artificial intelligence technology has made it possible to realize automatic evaluation systems for singing, and relevant research been able achieve accurate evaluations with respect pitch rhythm, but on singing-voice timbre remained at the level theoretical analysis. Timbre is closely related expression performance, breath control, emotional rendering, other aspects singing skills, a crucial impact song interpretation. purpose this investigate method timbre. At present...
The article presents the design and development of a virtual fretless Chinese stringed instrument App with Duxianqin as an example, whose performance is expected to be no different from real instrument. digital simulation musical instruments mainly divided into two parts: continuous pitch processing strings, sound produced by plucking strings. returns theory mechanics wave obtains quantitative relationship between string frequency its deformation elongation. selected in this instrument,...
Music generation has always been a hot topic. When discussing symbolic music, melody or harmonies are usually seen as the only generating targets. But in fact, playing techniques also quite an important part of music. In this paper, we discuss problem by seeing it tagging problem. We propose model that can use both current data and external knowledge. Experiments were carried out applying proposed Chinese bamboo flute results show our method make generated music more lively.
Since the 20th century, synthesizers and MIDI keyboards offer composers more convenient operation a higher creative ceiling. However, traditional Chinese plucked instruments like Pipa have only just begun to explore realms of electronic music automatic notation. This study focuses on as subject, aiming design implement device that captures performance-related data while maintaining original conditions instrument, automatically transcribes this into musical notation.This presents collection...
In this paper, we propose a method that takes the audio of solo pipa performance as input and utilizes pre-trained models other instruments to enhance pipa's 3D skeleton motion performance. This approach aims address limited availability training data. The key idea is believe all exhibit similar trends in their large limbs while playing. Therefore, can utilize videos playing pre-train capture limbs. Subsequently, fine-tune model using solo, thereby addressing issue data for solo. We...