2021/04/05 by Yuan Gong, Gong, Yuan, Yu-An Chung +3 · 198 citations
Computer Science · Engineering · #Artificial intelligence #Artificial neural network #Computer science #Convolution (computer science) #Convolutional neural network #Engineering #Music and Audio Processing #Pattern recognition (psychology) #Spectrogram #Speech Recognition and Synthesis #Speech and Audio Processing #Speech recognition #Transformer #cs.AI #cs.SD
paper · pdf · doi:10.48550/arxiv.2104.01778
published in arXiv (Cornell University) (Cornell University) · Accepted at Interspeech 2021. Code at https://github.com/YuanGongND/ast
openalex publication_date 2021/04/05 · arxiv created 2021/07/08 · arxiv updated 2021/07/12 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
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.