PANNs: Large-Scale Pretrained Audio Neural Networks for Audio Pattern Recognition

TL;DR

This paper introduces PANNs, large-scale pretrained audio neural networks trained on AudioSet, achieving state-of-the-art results in audio tagging and multiple related tasks, demonstrating the effectiveness of large-scale pretraining in audio pattern recognition.

Contribution

The paper presents the first large-scale pretrained audio neural networks (PANNs) trained on AudioSet, improving performance across various audio recognition tasks.

Findings

PANNs achieve a state-of-the-art mean average precision of 0.439 on AudioSet tagging.

PANNs outperform previous systems on several audio pattern recognition tasks.

The proposed Wavegram-Logmel-CNN architecture effectively utilizes both waveform and spectrogram inputs.

Abstract

Audio pattern recognition is an important research topic in the machine learning area, and includes several tasks such as audio tagging, acoustic scene classification, music classification, speech emotion classification and sound event detection. Recently, neural networks have been applied to tackle audio pattern recognition problems. However, previous systems are built on specific datasets with limited durations. Recently, in computer vision and natural language processing, systems pretrained on large-scale datasets have generalized well to several tasks. However, there is limited research on pretraining systems on large-scale datasets for audio pattern recognition. In this paper, we propose pretrained audio neural networks (PANNs) trained on the large-scale AudioSet dataset. These PANNs are transferred to other audio related tasks. We investigate the performance and computational complexity of PANNs modeled by a variety of convolutional neural networks. We propose an architecture called Wavegram-Logmel-CNN using both log-mel spectrogram and waveform as input feature. Our best PANN system achieves a state-of-the-art mean average precision (mAP) of 0.439 on AudioSet tagging, outperforming the best previous system of 0.392. We transfer PANNs to six audio pattern recognition tasks, and demonstrate state-of-the-art performance in several of those tasks. We have released the source code and pretrained models of PANNs: https://github.com/qiuqiangkong/audioset_tagging_cnn.