Multi-Modal Retrieval For Large Language Model Based Speech Recognition

TL;DR

This paper introduces multi-modal retrieval methods for speech recognition using external data, demonstrating significant improvements over text-only approaches and achieving state-of-the-art results on a question-answering dataset.

Contribution

It extends retrieval techniques to multi-modal data in speech recognition, combining kNN-LM and cross-attention methods for enhanced performance.

Findings

Speech-based multi-modal retrieval outperforms text-based retrieval.

Up to 50% improvement in word error rate.

Achieves state-of-the-art results on Spoken-Squad.

Abstract

Retrieval is a widely adopted approach for improving language models leveraging external information. As the field moves towards multi-modal large language models, 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 language model baseline. Furthermore, we achieve state-of-the-art recognition results on the Spoken-Squad question answering dataset.