Search arXiv⌕ Search
Home/Archive/arXiv

arXiv · 2506.08967

Step-Audio-AQAA: a Fully End-to-End Expressive Large Audio Language Model

Ailin Huang·Bingxin Li·Bruce Wang·Boyong Wu·Chao Yan·Chengli Feng·Heng Wang·Hongyu Zhou·Hongyuan Wang·Jingbei Li·Jianjian Sun·Joanna Wang·Mingrui Chen·Peng Liu·Ruihang Miao·Shilei Jiang·Tian Fei·Wang You·Xi Chen·Xuerui Yang·Yechang Huang·Yuxiang Zhang·Zheng Ge·Zheng Gong·Zhewei Huang·Zixin Zhang·Bin Wang·Bo Li·Buyun Ma·Changxin Miao·Changyi Wan·Chen Xu·Dapeng Shi·Dingyuan Hu·Enle Liu·Guanzhe Huang·Gulin Yan·Hanpeng Hu·Haonan Jia·Jiahao Gong·Jiaoren Wu·Jie Wu·Jie Yang·Junzhe Lin·Kaixiang Li·Lei Xia·Longlong Gu·Ming Li·Nie Hao·Ranchen Ming·Shaoliang Pang·Siqi Liu·Song Yuan·Tiancheng Cao·Wen Li·Wenqing He·Xu Zhao·Xuelin Zhang·Yanbo Yu·Yinmin Zhong·Yu Zhou·Yuanwei Liang·Yuanwei Lu·Yuxiang Yang·Zidong Yang·Zili Zhang·Binxing Jiao·Heung-Yeung Shum·Jiansheng Chen·Jing Li·Xiangyu Zhang·Xinhao Zhang·Yibo Zhu·Daxin Jiang·Shuchang Zhou·Chen Hu
View original source ↗Read source document ↗Export metadata ↓

Abstract

Large Audio-Language Models (LALMs) have significantly advanced intelligent human-computer interaction, yet their reliance on text-based outputs limits their ability to generate natural speech responses directly, hindering seamless audio interactions. To address this, we introduce Step-Audio-AQAA, a fully end-to-end LALM designed for Audio Query-Audio Answer (AQAA) tasks. The model integrates a dual-codebook audio tokenizer for linguistic and semantic feature extraction, a 130-billion-parameter backbone LLM and a neural vocoder for high-fidelity speech synthesis. Our post-training approach employs interleaved token-output of text and audio to enhance semantic coherence and combines Direct Preference Optimization (DPO) with model merge to improve performance. Evaluations on the StepEval-Audio-360 benchmark demonstrate that Step-Audio-AQAA excels especially in speech control, outperforming the state-of-art LALMs in key areas. This work contributes a promising solution for end-to-end LALMs and highlights the critical role of token-based vocoder in enhancing overall performance for AQAA tasks.

Explore related subjects

cs.SDcs.CLeess.AS

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Ailin Huang, Bingxin Li, Bruce Wang, Boyong Wu, Chao Yan, Chengli Feng, Heng Wang, Hongyu Zhou, Hongyuan Wang, Jingbei Li, Jianjian Sun, Joanna Wang, Mingrui Chen, Peng Liu, Ruihang Miao, Shilei Jiang, Tian Fei, Wang You, Xi Chen, Xuerui Yang, Yechang Huang, Yuxiang Zhang, Zheng Ge, Zheng Gong, Zhewei Huang, Zixin Zhang, Bin Wang, Bo Li, Buyun Ma, Changxin Miao, Changyi Wan, Chen Xu, Dapeng Shi, Dingyuan Hu, Enle Liu, Guanzhe Huang, Gulin Yan, Hanpeng Hu, Haonan Jia, Jiahao Gong, Jiaoren Wu, Jie Wu, Jie Yang, Junzhe Lin, Kaixiang Li, Lei Xia, Longlong Gu, Ming Li, Nie Hao, Ranchen Ming, Shaoliang Pang, Siqi Liu, Song Yuan, Tiancheng Cao, Wen Li, Wenqing He, Xu Zhao, Xuelin Zhang, Yanbo Yu, Yinmin Zhong, Yu Zhou, Yuanwei Liang, Yuanwei Lu, Yuxiang Yang, Zidong Yang, Zili Zhang, Binxing Jiao, Heung-Yeung Shum, Jiansheng Chen, Jing Li, Xiangyu Zhang, Xinhao Zhang, Yibo Zhu, Daxin Jiang, Shuchang Zhou, Chen Hu. 2025-06-13. Step-Audio-AQAA: a Fully End-to-End Expressive Large Audio Language Model. https://arxiv.org/abs/2506.08967

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Text-only adaptation in LLM-based ASR through text denoising

Adapting large language model (LLM)-based automatic speech recognition (ASR) systems to new domains using text-only data is a significant yet underexplored challenge. Standard fine-tuning of the LLM on the target domain text often disrupts the critical alignment between the speech and text modality learned by the projector, degrading performance. We introduce a novel text-only adaptation method that frames this process as a text denoising task. Our approach trains the LLM to recover clean transcripts from noisy inputs. This process effectively adapts the model to a target domain while preserving cross-modal alignment. Our solution is lightweight, requiring no architectural changes or additional parameters. Extensive evaluation on two datasets demonstrates up to 22.1% relative improvement, outperforming recent state-of-the-art text-only adaptation methods.

cs.SD↗
arXiv2026

Discrete vs. Continuous: A Comprehensive Study of Unified Audio Understanding in LALMs

Large Audio Language Models (LALMs) utilize either continuous features or discrete tokens, yet the optimal representation paradigm for general audio understanding remains debated. Existing benchmarks often focus on narrow domains or evaluate encoders outside LALM contexts. To address these gaps, we systematically evaluate continuous and discrete representations across speech, sound and music. Utilizing our UniARC framework with dual evaluation strategies across model scales from SmolLM2-135M to Llama-3-8B, we analyze the dynamic relationships of data volume, model capacity, and computational efficiency. Our results reveal the pivotal role of semantic constraints in tokenization for audio understanding and demonstrate that scaling backbones fail to compensate for information loss in audio representation, especially in data-limited tasks. These findings offer practical guidance for balancing semantic density, fidelity, and efficiency in future LALMs.

cs.SD↗
arXiv2026

Synthesis and editing of multi-instrument audio mixtures using scalar-quantised latents with MIDI Span conditioning

Music creation often involves iterative refinement, changing selected musical details while retaining the rest. To support such refinement, we introduce SpanSynth-Edit, a flow-matching model for MIDI-guided synthesis and editing of multi-instrument audio mixtures using low-frame-rate scalar-quantised latents. MIDI Span encodes instrument-labelled note lifecycles as unordered event sets with continuous-valued attributes and pools each set into one conditioning vector per audio-latent frame. The model uses contextual audio for instrument-specific timbre guidance and supports editing by resynthesising the target region from revised MIDI. Experiments on single- and multi-instrument benchmarks show competitive performance and demonstrate within-frame onset control. We also discuss limitations of transcription-based note-adherence evaluation.

cs.SD↗