Multi-Modal Retrieval For Large Language Model Based Speech Recognition

Abstract: Retrieval is a widely adopted approach for improving LLMs leveraging external information. As the field moves towards multi-modal LLMs, it is important to extend the pure text based methods to incorporate other modalities in retrieval as well for applications across the wide spectrum of machine learning tasks and data types. In this work, we propose multi-modal retrieval with two approaches: kNN-LM and cross-attention techniques. We demonstrate the effectiveness of our retrieval approaches empirically by applying them to automatic speech recognition tasks with access to external information. Under this setting, we show that speech-based multi-modal retrieval outperforms text based retrieval, and yields up to 50 % improvement in word error rate over the multi-modal LLM baseline. Furthermore, we achieve state-of-the-art recognition results on the Spoken-Squad question answering dataset.