Objective Evaluation of Guqin Timbre by Digital Signal Processing

Chen JingLei,Yu YiBiao,Wu JiHeng
DOI: https://doi.org/10.1109/icsp.2018.8652430
2018-01-01
Abstract:Guqin is a Chinese traditional musical instrument with history of 3000 years. The evaluation of Guqin timbre is mainly based on subjective evaluation at present, which is arbitrary and lack of persuasion. Based on the quantitative acoustic signal analysis and pattern recognition theory, this paper proposes an objective evaluation method of Guqin timbre. To analyse acoustic characteristics of Guqin music signal, the Mel-Frequency Cepstral Coefficient (MFCC) is extracted according to the human auditory system principle, and the Guqin Timbre Model (GTM) was established by statistical study. In the evaluation, based on the standard method, the Bayesian likelihood is calculated to measure the similarity between the evaluated Guqin and the reference Guqin. The experimental results show that the GTM can objectively describe the timbre characteristics of Guqin, and the objective evaluation results are in good agreement with the subjective evaluation results, indicating the validity of the Guqin timbre evaluation method proposed in this paper.
What problem does this paper attempt to address?