An Encoder-Decoder Based Audio Captioning System With Transfer and Reinforcement Learning

TL;DR

This paper introduces an encoder-decoder audio captioning system enhanced with transfer learning and reinforcement learning, improving evaluation metrics but with mixed effects on caption quality.

Contribution

The study proposes a novel encoder-decoder architecture for audio captioning that incorporates transfer learning and reinforcement learning to address data scarcity and metric optimization.

Findings

System ranked 3rd in DCASE 2021 Task 6

Transfer learning significantly improves performance

Reinforcement learning enhances metric scores but may affect caption quality

Abstract

Automated audio captioning aims to use natural language to describe the content of audio data. This paper presents an audio captioning system with an encoder-decoder architecture, where the decoder predicts words based on audio features extracted by the encoder. To improve the proposed system, transfer learning from either an upstream audio-related task or a large in-domain dataset is introduced to mitigate the problem induced by data scarcity. Besides, evaluation metrics are incorporated into the optimization of the model with reinforcement learning, which helps address the problem of ``exposure bias'' induced by ``teacher forcing'' training strategy and the mismatch between the evaluation metrics and the loss function. The resulting system was ranked 3rd in DCASE 2021 Task 6. Ablation studies are carried out to investigate how much each element in the proposed system can contribute to final performance. The results show that the proposed techniques significantly improve the scores of the evaluation metrics, however, reinforcement learning may impact adversely on the quality of the generated captions.