Audio Retrieval with Natural Language Queries
This work addresses a limitedly studied problem in cross-modal retrieval for audio and text, but is incremental as it focuses on benchmarking and baseline methods.
The paper tackles the problem of retrieving audio using free-form natural language queries by introducing new benchmarks from Audiocaps and Clotho datasets, and establishes baselines showing benefits from pre-training on diverse audio tasks.
We consider the task of retrieving audio using free-form natural language queries. To study this problem, which has received limited attention in the existing literature, we introduce challenging new benchmarks for text-based audio retrieval using text annotations sourced from the Audiocaps and Clotho datasets. We then employ these benchmarks to establish baselines for cross-modal audio retrieval, where we demonstrate the benefits of pre-training on diverse audio tasks. We hope that our benchmarks will inspire further research into cross-modal text-based audio retrieval with free-form text queries.