# Aligning Spoken Dialogue Models from User Interactions

arXiv: https://arxiv.org/abs/2506.21463 · HTML: https://mlsanity.com/paper/2506.21463/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Placed in Preference alignment for recommenders and RAG, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 39 of its 46 references that are in the map.
- First posted on arXiv 26 Jun 2025.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Preference alignment for recommenders and RAG](https://mlsanity.com/alignment/preference-alignment-for-recommenders-and-rag/)

## Builds on (most cited first, 30 of 39)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension](https://mlsanity.com/paper/1705.03551/) (2017)
- [Fine-Tuning Language Models from Human Preferences](https://mlsanity.com/paper/1909.08593/) (2019)
- [Tulu 3: Pushing Frontiers in Open Language Model Post-Training](https://mlsanity.com/paper/2411.15124/) (2024)
- [SimPO: Simple Preference Optimization with a Reference-Free Reward](https://mlsanity.com/paper/2405.14734/) (2024)
- [SoundStream: An End-to-End Neural Audio Codec](https://mlsanity.com/paper/2107.03312/) (2021)
- [High Fidelity Neural Audio Compression](https://mlsanity.com/paper/2210.13438/) (2022)
- [RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedback](https://mlsanity.com/paper/2309.00267/) (2023)
- [AudioLM: a Language Modeling Approach to Audio Generation](https://mlsanity.com/paper/2209.03143/) (2022)
- [Moshi: a speech-text foundation model for real-time dialogue](https://mlsanity.com/paper/2410.00037/) (2024)
- [SpeechGPT: Empowering Large Language Models with Intrinsic Cross-Modal Conversational Abilities](https://mlsanity.com/paper/2305.11000/) (2023)
- [XSTest: A Test Suite for Identifying Exaggerated Safety Behaviours in Large Language Models](https://mlsanity.com/paper/2308.01263/) (2023)
- [WildGuard: Open One-Stop Moderation Tools for Safety Risks, Jailbreaks, and Refusals of LLMs](https://mlsanity.com/paper/2406.18495/) (2024)
- [Interpretable Preferences via Multi-Objective Reward Modeling and Mixture-of-Experts](https://mlsanity.com/paper/2406.12845/) (2024)
- [HelpSteer2: Open-source dataset for training top-performing reward models](https://mlsanity.com/paper/2406.08673/) (2024)
- [Scaling Laws for Reward Model Overoptimization in Direct Alignment Algorithms](https://mlsanity.com/paper/2406.02900/) (2024)
- [What makes a good conversation? How controllable attributes affect human judgments](https://mlsanity.com/paper/1902.08654/) (2019)
- [UniAudio: An Audio Foundation Model Toward Universal Audio Generation](https://mlsanity.com/paper/2310.00704/) (2023)
- [Generative Spoken Dialogue Language Modeling](https://mlsanity.com/paper/2203.16502/) (2022)
- [ALERT: A Comprehensive Benchmark for Assessing Large Language Models' Safety through Red Teaming](https://mlsanity.com/paper/2404.08676/) (2024)
- [Spoken Question Answering and Speech Continuation Using Spectrogram-Powered LLM](https://mlsanity.com/paper/2305.15255/) (2023)
- [Human Evaluation of Conversations is an Open Problem: comparing the sensitivity of various methods for evaluating dialogue agents](https://mlsanity.com/paper/2201.04723/) (2022)
- [SpeechAlign: Aligning Speech Generation to Human Preferences](https://mlsanity.com/paper/2404.05600/) (2024)
- [Scaling Speech-Text Pre-training with Synthetic Interleaved Data](https://mlsanity.com/paper/2411.17607/) (2024)

## Built on by (most cited first, 11 of 11)

- [Game-Time: Evaluating Temporal Dynamics in Spoken Language Models](https://mlsanity.com/paper/2509.26388/) (2025)
- [Multi-Faceted Interactivity Alignment in Full-Duplex Speech Models](https://mlsanity.com/paper/2606.11167/) (2026)
- [Optimizing Conversational Quality in Spoken Dialogue Systems with Reinforcement Learning from AI Feedback](https://mlsanity.com/paper/2601.19063/) (2026)
- [ASPIRin: Action Space Projection for Interactivity-Optimized Reinforcement Learning in Full-Duplex Speech Language Models](https://mlsanity.com/paper/2604.10065/) (2026)
- [WavAlign: Enhancing Intelligence and Expressiveness in Spoken Dialogue Models via Adaptive Hybrid Post-Training](https://mlsanity.com/paper/2604.14932/) (2026)
- [TurnGuide: Enhancing Meaningful Full Duplex Spoken Interactions via Dynamic Turn-Level Text-Speech Interleaving](https://mlsanity.com/paper/2508.07375/) (2025)
- [Dual-Axis Generative Reward Model Toward Semantic and Turn-taking Robustness in Interactive Spoken Dialogue Models](https://mlsanity.com/paper/2604.14920/) (2026)
- [RetroThinker: Enabling Retrospective Thinking in Speech LLMs](https://mlsanity.com/paper/2609.11864/) (2026)
- [Decoupling Strategy and Execution in Task-Focused Dialogue via Goal-Oriented Preference Optimization](https://mlsanity.com/paper/2602.15854/) (2026)
- [NaturalFlow: Reducing Disruptive Pauses for Natural Speech Flow in Simultaneous Speech-to-Speech Translation](https://mlsanity.com/paper/2606.13121/) (2026)
- [SteerDuplex: Steerable Duplex Speech Dialogue Models](https://mlsanity.com/paper/2609.12623/) (2026)

## Abstract (arXiv)

We propose a novel preference alignment framework for improving spoken dialogue models on real-time conversations from user interactions. Current preference learning methods primarily focus on text-based language models, and are not directly suited to the complexities of real-time speech interactions, with richer dynamics (e.g. interruption, interjection) and no explicit segmentation between speaker turns.We create a large-scale dataset of more than 150,000 preference pairs from raw multi-turn speech conversations, annotated with AI feedback, to cover preferences over both linguistic content and temporal context variations. We leverage offline alignment methods to finetune a full-duplex autoregressive speech-to-speech model. Extensive experiments demonstrate that feedback on generic conversations can be consistently effective in improving spoken dialogue models to produce more factual, safer and more contextually aligned interactions. We deploy the finetuned model and conduct holistic human evaluations to assess the impact beyond single-turn conversations. Our findings shed light on the importance of a well-calibrated balance among various dynamics, crucial for natural real-time speech dialogue systems.