Score-informed syllable segmentation for a cappella singing voice with convolutional neural networks
This addresses a domain-specific challenge in music information retrieval for jingju singing, with incremental improvements in segmentation techniques.
The paper tackles the problem of segmenting jingju a cappella singing phrases into syllables by proposing a score-informed method using convolutional neural networks and a Viterbi algorithm, which outperforms the state-of-the-art in segmentation accuracy.
This paper introduces a new score-informed method for the segmentation of jingju a cappella singing phrase into syllables. The proposed method estimates the most likely sequence of syllable boundaries given the estimated syllable onset detection function (ODF) and its score. Throughout the paper, we first examine the jingju syllables structure and propose a definition of the term "syllable onset". Then, we identify which are the challenges that jingju a cappella singing poses. Further, we investigate how to improve the syllable ODF estimation with convolutional neural networks (CNNs). We propose a novel CNN architecture that allows to efficiently capture different time-frequency scales for estimating syllable onsets. In addition, we propose using a score-informed Viterbi algorithm -instead of thresholding the onset function-, because the available musical knowledge we have (the score) can be used to inform the Viterbi algorithm in order to overcome the identified challenges. The proposed method outperforms the state-of-the-art in syllable segmentation for jingju a cappella singing. We further provide an analysis of the segmentation errors which points possible research directions.