Transformer Based Unsupervised Pre-Training for Acoustic Representation Learning

• Published in: Proceedings of the ... IEEE International Conference on Acoustics, Speech and Signal Processing (1998) pp. 6933 - 6937
• Main Authors: Zhang, Ruixiong, Wu, Haiwei, Li, Wubo, Jiang, Dongwei, Zou, Wei, Li, Xiangang
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.01.2021
• Subjects: acoustic representation learning; Acoustics; Conferences; Emotion recognition; Event detection; Signal processing; Speech recognition; Training data; Transformer; unsupervised pre-training
• DOI: 10.1109/ICASSP39728.2021.9414996

Recently, a variety of acoustic tasks and related applications arised. For many acoustic tasks, the labeled data size may be limited. To handle this problem, we propose an unsupervised pre-training method using Transformer based encoder to learn a general and robust high-level representation for all acoustic tasks. Experiments have been conducted on three kinds of acoustic tasks: speech emotion recognition, sound event detection and speech translation. All the experiments have shown that pre-training using its own training data can significantly improve the performance. With a larger pre-training data combining MuST-C, Librispeech and ESC-US datasets, for speech emotion recognition, the UAR can further improve absolutely 4.3% on IEMOCAP dataset. For sound event detection, the F1 score can further improve absolutely 1.5% on DCASE2018 task5 development set and 2.1% on evaluation set. For speech translation, the BLEU score can further improve relatively 12.2% on En-De dataset and 8.4% on En-Fr dataset.