Align-slm: Textless Spoken Language Models With Reinforcement Learning From AI Feedback

Abstract

While textless Spoken Language Models (SLMs) have shown potential in end-to-end speech-to-speech modeling, they still lag behind text-based Large Language Models (LLMs) in terms of semantic coherence and relevance. This work introduces the Align-SLM framework, which leverages preference optimization inspired by Reinforcement Learning with AI Feedback (RLAIF) to enhance the semantic understanding of SLMs. Our approach generates multiple speech continuations from a given prompt and uses semantic metrics to create preference data for Direct Preference Optimization (DPO). We evaluate the framework using ZeroSpeech 2021 benchmarks for lexical and syntactic modeling, the spoken version of the StoryCloze dataset for semantic coherence, and other speech generation metrics, including the GPT4-o score and human evaluation. Experimental results show that our method achieves state-of-the-art performance for SLMs on most benchmarks, highlighting the importance of preference optimization to improve

Stats

• citations8
• S2 citations—
• github stars0
• HF likes0
• heat score7.16
• arxiv keylin2024align

Related papers

• SLIDE: Integrating Speech Language Model With LLM For Spontaneous Spoken Dialogue Generation (2025)2.26
• Desta: Enhancing Speech Language Models Through Descriptive Speech-text Alignment (2024)9.03
• Speech Recognition With Llms Adapted To Disordered Speech Using Reinforcement Learning (2024)5.24
• BLSP: Bootstrapping Language-speech Pre-training Via Behavior Alignment Of Continuation Writing (2023)0.00
• Exploring Fine-tuning Of Large Audio Language Models For Spoken Language Understanding Under Limited Speech Data (2025)0.00
• PSLM: Parallel Generation Of Text And Speech With Llms For Low-latency Spoken Dialogue Systems (2024)2.26
• Desta2: Developing Instruction-following Speech Language Model Without Speech Instruction-tuning Data (2024)8.82
• Boosting Large Language Model For Speech Synthesis: An Empirical Study (2023)6.77