FAST: Fast Audio Spectrogram Transformer

TL;DR

FAST is a lightweight, efficient audio spectrogram transformer that combines CNNs and transformers, achieving state-of-the-art results in real-time audio classification with significantly fewer parameters.

Contribution

Introduces FAST, a novel architecture combining CNNs and transformers with Lipschitz attention for efficient, stable, real-time audio classification.

Findings

Achieves state-of-the-art performance on ADIMA and AudioSet datasets.

Uses up to 150x fewer parameters than existing models.

Demonstrates effectiveness in multilingual profanity and abuse detection.

Abstract

In audio classification, developing efficient and robust models is critical for real-time applications. Inspired by the design principles of MobileViT, we present FAST (Fast Audio Spectrogram Transformer), a new architecture that combines convolutional neural networks (CNNs) and transformers to capitalize on the strengths of both. FAST integrates the local feature extraction efficiencies of CNNs with the global context modeling capabilities of transformers, resulting in a model that is powerful yet lightweight, well-suited to a real-time or mobile use case. Additionally, we incorporate Lipschitz continuous attention mechanisms to improve training stability and accelerate convergence. We evaluate FAST on the ADIMA dataset, a multilingual corpus towards real-time profanity and abuse detection, as well as on the more traditional AudioSet. Our results show that FAST achieves state-of-the-art performance on both the ADIMA and AudioSet classification tasks and in some cases surpasses existing benchmarks while using up to 150x fewer parameters.