MAE-AST: Masked Autoencoding Audio Spectrogram Transformer

Autor:	Baade, Alan, Peng, Puyuan, Harwath, David
Rok vydání:	2022
Předmět:	Electrical Engineering and Systems Science - Audio and Speech Processing Computer Science - Artificial Intelligence Computer Science - Computation and Language Computer Science - Machine Learning Computer Science - Sound
Druh dokumentu:	Working Paper
Popis:	In this paper, we propose a simple yet powerful improvement over the recent Self-Supervised Audio Spectrogram Transformer (SSAST) model for speech and audio classification. Specifically, we leverage the insight that the SSAST uses a very high masking ratio (75%) during pretraining, meaning that the vast majority of self-attention compute is performed on mask tokens. We address this by integrating the encoder-decoder architecture from Masked Autoencoders are Scalable Vision Learners (MAE) into the SSAST, where a deep encoder operates on only unmasked input, and a shallow decoder operates on encoder outputs and mask tokens. We find that MAE-like pretraining can provide a 3x speedup and 2x memory usage reduction over the vanilla SSAST using current audio pretraining strategies with ordinary model and input sizes. When fine-tuning on downstream tasks, which only uses the encoder, we find that our approach outperforms the SSAST on a variety of downstream tasks. We further conduct comprehensive evaluations into different strategies of pretraining and explore differences in MAE-style pretraining between the visual and audio domains. Comment: Submitted to INTERSPEECH. 5 pages, 2 figures, 5 tables
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2203.16691 Zobrazit plný text záznamu View this record from Arxiv