Toward Leveraging Pre-Trained Self-Supervised Frontends for Automatic Singing Voice Understanding Tasks: Three Case Studies
Author:
- 31 October 2023
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1745-1752
Abstract
Automatic singing voice understanding tasks, such as singer identification, singing voice transcription, and singing technique classification, benefit from data-driven approaches that utilize deep learning techniques. These approaches work well even under the rich diversity of vocal and noisy samples owing to their representation ability. However, the limited availability of labeled data remains a significant obstacle to achieving satisfactory performance. In recent years, self-supervised learning models (SSL models) have been trained using large amounts of unlabeled data in the field of speech processing and music classification. By fine-tuning these models for the target tasks, comparable performance to conventional supervised learning can be achieved with limited training data. Therefore, in this paper, we investigate the effectiveness of SSL models for various singing voice recognition tasks. We report the results of experiments comparing SSL models for three different tasks (i.e., singer identification, singing voice transcription, and singing technique classification) as initial exploration and aim to discuss these findings. Experimental results show that each SSL model achieves comparable performance and sometimes outperforms compared to state-of-the-art methods on each task. We also conducted a layer-wise analysis to further understand the behavior of the SSL models.Keywords
All Related Versions
Funding Information
- National Taiwan University
This publication has 30 references indexed in Scilit:
- Automatic Lyrics Transcription using Dilated Convolutional Neural Networks with Self-Attention2020
- Addressing The Confounds Of Accompaniments In Singer Identification2020
- Music Artist Classification with Convolutional Recurrent Neural Networks2019
- An Introduction to Signal Processing for Singing-Voice Analysis: High Notes in the Effort to Automate the Understanding of Vocals in MusicIEEE Signal Processing Magazine, 2018
- Towards the characterization of singing styles in world music2017
- CNN architectures for large-scale audio classification2017
- Librispeech: An ASR corpus based on public domain audio books2015
- On the Properties of Neural Machine Translation: Encoder–Decoder Approaches2014
- MiruSinger: A Singing Skill Visualization Interface Using Real-Time Feedback and Music CD Recordings as Referential Data2007
- Query by humming1995