PSLM: Parallel Generation of Text and Speech with LLMs for Low-Latency Spoken Dialogue Systems

TL;DR

This paper introduces PSLM, a multimodal language model that enables parallel generation of text and speech to reduce response latency in spoken dialogue systems, demonstrating improved speed without sacrificing quality.

Contribution

The study extends language models to support simultaneous text and speech generation, addressing latency issues in spoken dialogue systems.

Findings

Improved response latency in spoken question answering tasks.

Maintained response quality despite reduced latency.

Further latency reduction by generating speech in multiple sequences.

Abstract

Multimodal language models that process both text and speech have a potential for applications in spoken dialogue systems. However, current models face two major challenges in response generation latency: (1) generating a spoken response requires the prior generation of a written response, and (2) speech sequences are significantly longer than text sequences. This study addresses these issues by extending the input and output sequences of the language model to support the parallel generation of text and speech. Our experiments on spoken question answering tasks demonstrate that our approach improves latency while maintaining the quality of response content. Additionally, we show that latency can be further reduced by generating speech in multiple sequences. Demo samples are available at https://rinnakk.github.io/research/publications/PSLM.