Residual Attention Based Network for Automatic Classification of Phonation Modes
This work addresses a specific problem for music information retrieval researchers and practitioners, offering an incremental improvement in classification accuracy for phonation modes.
The paper tackled the problem of automatically classifying phonation modes in singing, proposing a Residual Attention network that achieved a highest classification accuracy of 94.58%, which is 2.29% higher than the baseline.
Phonation mode is an essential characteristic of singing style as well as an important expression of performance. It can be classified into four categories, called neutral, breathy, pressed and flow. Previous studies used voice quality features and feature engineering for classification. While deep learning has achieved significant progress in other fields of music information retrieval (MIR), there are few attempts in the classification of phonation modes. In this study, a Residual Attention based network is proposed for automatic classification of phonation modes. The network consists of a convolutional network performing feature processing and a soft mask branch enabling the network focus on a specific area. In comparison experiments, the models with proposed network achieve better results in three of the four datasets than previous works, among which the highest classification accuracy is 94.58%, 2.29% higher than the baseline.