PSLM: Parallel Generation Of Text And Speech With Llms For Low-latency Spoken Dialogue Systems

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.

Stats

• citations1
• S2 citations—
• github stars0
• HF likes0
• heat score2.26
• arxiv keymitsui2024pslm

Related papers

• SLIDE: Integrating Speech Language Model With LLM For Spontaneous Spoken Dialogue Generation (2025)2.26
• Paralinguistics-enhanced Large Language Modeling Of Spoken Dialogue (2023)0.00
• Get Large Language Models Ready To Speak: A Late-fusion Approach For Speech Generation (2024)5.24
• Towards Efficient Speech-text Jointly Decoding Within One Speech Language Model (2025)0.00
• SLM-S2ST: A Multimodal Language Model For Direct Speech-to-speech Translation (2025)0.00
• Recent Advances In Speech Language Models: A Survey (2024)14.64
• Audiochatllama: Towards General-purpose Speech Abilities For Llms (2023)9.41
• Align-slm: Textless Spoken Language Models With Reinforcement Learning From AI Feedback (2024)7.16