AST: Audio Spectrogram Transformer
TL
TL;DR
The Audio Spectrogram Transformer is introduced, the first convolution-free, purely attention-based model for audio classification, which achieves new state-of-the-art results on various audio classification benchmarks.
Abstract
In the past decade, convolutional neural networks (CNNs) have been widely adopted as the main building block for end-to-end audio classification models, which aim to learn a direct mapping from audio spectrograms to corresponding labels. To better capture long-range global context, a recent trend is to add a self-attention mechanism on top of the CNN, forming a CNN-attention hybrid model. However, it is unclear whether the reliance on a CNN is necessary, and if neural networks purely based on attention are sufficient to obtain good performance in audio classification. In this paper, we answer the question by introducing the Audio Spectrogram Transformer (AST), the first convolution-free, purely attention-based model for audio classification. We evaluate AST on various audio classification benchmarks, where it achieves new state-of-the-art results of 0.485 mAP on AudioSet, 95.6% accuracy on ESC-50, and 98.1% accuracy on Speech Commands V2.
References35 items
1
Tokens-to-Token ViT: Training Vision Transformers from Scratch on ImageNet
2
Training data-efficient image transformers & distillation through attention
3
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
4
Rethinking CNN Models for Audio Classification
5