Improving Audio Caption Fluency with Automatic Error Correction
This addresses a specific error issue in audio captioning for applications requiring fluent descriptions, but it is incremental as it builds on existing models.
The paper tackled the problem of false-repetition errors in automated audio captioning by proposing a post-processing error correction method, which significantly improved fluency on two benchmark datasets while preserving semantics.
Automated audio captioning (AAC) is an important cross-modality translation task, aiming at generating descriptions for audio clips. However, captions generated by previous AAC models have faced ``false-repetition'' errors due to the training objective. In such scenarios, we propose a new task of AAC error correction and hope to reduce such errors by post-processing AAC outputs. To tackle this problem, we use observation-based rules to corrupt captions without errors, for pseudo grammatically-erroneous sentence generation. One pair of corrupted and clean sentences can thus be used for training. We train a neural network-based model on the synthetic error dataset and apply the model to correct real errors in AAC outputs. Results on two benchmark datasets indicate that our approach significantly improves fluency while maintaining semantic information.