Heyang Liu

Zihan Zhao, Yiyang Jiang, Heyang Liu et al. · cambridge

While Large Language Models (LLMs) have demonstrated commendable performance across a myriad of domains and tasks, existing LLMs still exhibit a palpable deficit in handling multimodal functionalities, especially for the Spoken Question Answering (SQA) task which necessitates precise alignment and deep interaction between speech and text features. To address the SQA challenge on LLMs, we initially curated the free-form and open-ended LibriSQA dataset from Librispeech, comprising Part I with natural conversational formats and Part II encompassing multiple-choice questions followed by answers and analytical segments. Both parts collectively include 107k SQA pairs that cover various topics. Given the evident paucity of existing speech-text LLMs, we propose a lightweight, end-to-end framework to execute the SQA task on the LibriSQA, witnessing significant results. By reforming ASR into the SQA format, we further substantiate our framework's capability in handling ASR tasks. Our empirical findings bolster the LLMs' aptitude for aligning and comprehending multimodal information, paving the way for the development of universal multimodal LLMs. The dataset and demo can be found at https://github.com/ZihanZhaoSJTU/LibriSQA.

Heyang Liu, Ziyang Cheng, Jiayi Huang et al.

Recent advances in Speech Large Language Models (Speech LLMs) have significantly enhanced spoken language understanding and reasoning. However, their contextual awareness is limited, struggling to perform speech recognition that effectively reflects the speaker's intent and topical context. In this paper, we propose LaSR (Latent Speech Reasoning), a novel training paradigm featuring a context-aware reasoning trajectory that leverages the latent reasoning process. Instead of generating explicit intermediate tokens, LaSR aligns chain-of-thought (CoT) supervision around the acoustic feature region of the targeted word, and introduces latent reasoning periods for context information grounding and transcriptional transition. Furthermore, to effectively benchmark contextual recognition on specialized vocabulary, we propose Spoken Darwin-Science, a large-scale corpus focusing on academic terminologies. Preliminary experiments on Fun-Audio-Chat demonstrate that LaSR significantly improves terminology recognition without introducing additional latency and consistently outperforms standard supervised fine-tuning baselines. Our findings highlight the potential of latent reasoning in building efficient, context-aware speech assistants.

Hongcheng Liu, Yusheng Liao, Siqv Ou et al.

The application of the Multi-modal Large Language Models (MLLMs) in medical clinical scenarios remains underexplored. Previous benchmarks only focus on the capacity of the MLLMs in medical visual question-answering (VQA) or report generation and fail to assess the performance of the MLLMs on complex clinical multi-modal tasks. In this paper, we propose a novel Medical Personalized Multi-modal Consultation (Med-PMC) paradigm to evaluate the clinical capacity of the MLLMs. Med-PMC builds a simulated clinical environment where the MLLMs are required to interact with a patient simulator to complete the multi-modal information-gathering and decision-making task. Specifically, the patient simulator is decorated with personalized actors to simulate diverse patients in real scenarios. We conduct extensive experiments to access 12 types of MLLMs, providing a comprehensive view of the MLLMs' clinical performance. We found that current MLLMs fail to gather multimodal information and show potential bias in the decision-making task when consulted with the personalized patient simulators. Further analysis demonstrates the effectiveness of Med-PMC, showing the potential to guide the development of robust and reliable clinical MLLMs. Code and data are available at https://github.com/LiuHC0428/Med-PMC.

Heyang Liu, Ziyang Cheng, Yuhao Wang et al.

The development of multi-modal large language models (LLMs) leads to intelligent approaches capable of speech interactions. As one of the most widely spoken languages globally, Mandarin is supported by most models to enhance their applicability and reach. However, the scarcity of comprehensive speech-to-speech (S2S) benchmarks in Mandarin contexts impedes systematic evaluation for developers and hinders fair model comparison for users. In this work, we propose VocalBench-zh, an ability-level divided evaluation suite adapted to Mandarin context consisting of 10 well-crafted subsets and over 10K high-quality instances, covering 12 user-oriented characters. The evaluation experiment on 14 mainstream models reveals the common challenges for current routes, and highlights the need for new insights into next-generation speech interactive systems. The evaluation codes and datasets will be available at https://github.com/SJTU-OmniAgent/VocalBench-zh.

Yuhao Wang, Yusheng Liao, Heyang Liu et al.

Recent advancements in Multimodal Large Language Models (MLLMs) have demonstrated exceptional capabilities in visual perception and understanding. However, these models also suffer from hallucinations, which limit their reliability as AI systems. We believe that these hallucinations are partially due to the models' struggle with understanding what they can and cannot perceive from images, a capability we refer to as self-awareness in perception. Despite its importance, this aspect of MLLMs has been overlooked in prior studies. In this paper, we aim to define and evaluate the self-awareness of MLLMs in perception. To do this, we first introduce the knowledge quadrant in perception, which helps define what MLLMs know and do not know about images. Using this framework, we propose a novel benchmark, the Self-Awareness in Perception for MLLMs (MM-SAP), specifically designed to assess this capability. We apply MM-SAP to a variety of popular MLLMs, offering a comprehensive analysis of their self-awareness and providing detailed insights. The experiment results reveal that current MLLMs possess limited self-awareness capabilities, pointing to a crucial area for future advancement in the development of trustworthy MLLMs. Code and data are available at https://github.com/YHWmz/MM-SAP.

Yuhao Wang, Heyang Liu, Ziyang Cheng et al.

Speech large language models (LLMs) have emerged as a prominent research focus in speech processing. We introduce VocalNet-1B and VocalNet-8B, a series of high-performance, low-latency speech LLMs enabled by a scalable and model-agnostic training framework designed for real-time voice interaction. Central to our contribution is the first application of multi-token prediction (MTP) to speech LLMs. This approach represents a paradigm shift from standard next-token prediction (NTP), offering simultaneous improvements in generation speed and quality. Informed by analysis of MTP's effect on speech generation and experimental comparisons, we designed a straightforward and highly effective MTP implementation. Experiments demonstrate that VocalNet performs on par with mainstream Omni LLMs even with limited training data, and significantly surpasses existing open-source speech LLMs. To foster reproducibility and community advancement, all model weights, inference code, training data, and framework implementations have been made publicly available at https://github.com/SJTU-OmniAgent/VocalNet

Yu Wang, Heyang Liu, Yuhao Wang et al.

Language, as an information medium created by advanced organisms, has always been a concern of neuroscience regarding how it is represented in the brain. Decoding linguistic representations in the evoked brain has shown groundbreaking achievements, thanks to the rapid improvement of neuroimaging, medical technology, life sciences and artificial intelligence. In this work, we present a taxonomy of brain-to-language decoding of both textual and speech formats. This work integrates two types of research: neuroscience focusing on language understanding and deep learning-based brain decoding. Generating discernible language information from brain activity could not only help those with limited articulation, especially amyotrophic lateral sclerosis (ALS) patients but also open up a new way for the next generation's brain-computer interface (BCI). This article will help brain scientists and deep-learning researchers to gain a bird's eye view of fine-grained language perception, and thus facilitate their further investigation and research of neural process and language decoding.

Ziyang Cheng, Yuhao Wang, Heyang Liu et al.

Recent Speech Large Language Models~(LLMs) have achieved impressive capabilities in end-to-end speech interaction. However, the prevailing autoregressive paradigm imposes strict serial constraints, limiting generation efficiency and introducing exposure bias. In this paper, we investigate Masked Diffusion Modeling~(MDM) as a non-autoregressive paradigm for speech LLMs and introduce VocalNet-MDM. To adapt MDM for streaming speech interaction, we address two critical challenges: training-inference mismatch and iterative overhead. We propose Hierarchical Block-wise Masking to align training objectives with the progressive masked states encountered during block diffusion decoding, and Iterative Self-Distillation to compress multi-step refinement into fewer steps for low-latency inference. Trained on a limited scale of only 6K hours of speech data, VocalNet-MDM achieves a 3.7$\times$--10$\times$ decoding speedup and reduces first-chunk latency by 34\% compared to AR baselines. It maintains competitive recognition accuracy while achieving state-of-the-art text quality and speech naturalness, demonstrating that MDM is a promising and scalable alternative for low-latency, efficient speech LLMs.

Yuhao Wang, Ziyang Cheng, Heyang Liu et al.

Current end-to-end spoken language models (SLMs) have made notable progress, yet they still encounter considerable response latency. This delay primarily arises from the autoregressive generation of speech tokens and the reliance on complex flow-matching models for speech synthesis. To overcome this, we introduce VocalNet-M2, a novel low-latency SLM that integrates a multi-codebook tokenizer and a multi-token prediction (MTP) strategy. Our model directly generates multi-codebook speech tokens, thus eliminating the need for a latency-inducing flow-matching model. Furthermore, our MTP strategy enhances generation efficiency and improves overall performance. Extensive experiments demonstrate that VocalNet-M2 achieves a substantial reduction in first chunk latency (from approximately 725ms to 350ms) while maintaining competitive performance across mainstream SLMs. This work also provides a comprehensive comparison of single-codebook and multi-codebook strategies, offering valuable insights for developing efficient and high-performance SLMs for real-time interactive applications.

Zhe Chen, Heyang Liu, Wenyi Yu et al.

Publishing open-source academic video recordings is an emergent and prevalent approach to sharing knowledge online. Such videos carry rich multimodal information including speech, the facial and body movements of the speakers, as well as the texts and pictures in the slides and possibly even the papers. Although multiple academic video datasets have been constructed and released, few of them support both multimodal content recognition and understanding tasks, which is partially due to the lack of high-quality human annotations. In this paper, we propose a novel multimodal, multigenre, and multipurpose audio-visual academic lecture dataset (M$^3$AV), which has almost 367 hours of videos from five sources covering computer science, mathematics, and medical and biology topics. With high-quality human annotations of the slide text and spoken words, in particular high-valued name entities, the dataset can be used for multiple audio-visual recognition and understanding tasks. Evaluations performed on contextual speech recognition, speech synthesis, and slide and script generation tasks demonstrate that the diversity of M$^3$AV makes it a challenging dataset.

Heyang Liu, Yuhao Wang, Ziyang Cheng et al.

The advancement of multimodal large language models has accelerated the development of speech-to-speech interaction systems. While natural monolingual interaction has been achieved, we find existing models exhibit deficiencies in language alignment. In our proposed Code-Switching Speech-to-Speech Benchmark (CS3-Bench), experiments on 7 mainstream models demonstrate a relative performance drop of up to 66% in knowledge-intensive question answering and varying degrees of misunderstanding in open-ended conversations. Starting from a model with severe performance deterioration, we propose both data constructions and training approaches to improve the language alignment capabilities, specifically employing Chain of Recognition (CoR) to enhance understanding and Keyword Highlighting (KH) to guide generation. Our approach improves the knowledge accuracy from 25.14% to 46.13%, with open-ended understanding rate from 64.5% to 86.5%, and significantly reduces pronunciation errors in the secondary language. CS3-Bench is available at https://huggingface.co/datasets/VocalNet/CS3-Bench.

Sheng Feng, Heyang Liu, Yu Wang et al.

In this paper, we introduce a groundbreaking end-to-end (E2E) framework for decoding invasive brain signals, marking a significant advancement in the field of speech neuroprosthesis. Our methodology leverages the comprehensive reasoning abilities of large language models (LLMs) to facilitate direct decoding. By fully integrating LLMs, we achieve results comparable to the state-of-the-art cascade models. Our findings underscore the immense potential of E2E frameworks in speech neuroprosthesis, particularly as the technology behind brain-computer interfaces (BCIs) and the availability of relevant datasets continue to evolve. This work not only showcases the efficacy of combining LLMs with E2E decoding for enhancing speech neuroprosthesis but also sets a new direction for future research in BCI applications, underscoring the impact of LLMs in decoding complex neural signals for communication restoration. Code will be made available at https://github.com/FsFrancis15/BrainLLM.

Heyang Liu, Yuhao Wang, Ziyang Cheng et al.

The rapid advancement of large language models (LLMs) has accelerated the development of multimodal models capable of speech communications. Unlike text interactions, speech conveys diverse information, including acoustic variations, paralanguage cues, and environmental context. However, existing evaluations of speech interaction models lack instances mimicking real scenarios and predominantly focus on the quality of their textual responses, overlooking critical aspects of vocal performance. To address this gap, we propose VocalBench, a comprehensive benchmark to assess the speech conversational abilities, comprising 9,400 carefully curated instances across four key dimensions: semantic quality, acoustic performance, conversational abilities, and robustness. It covers a broad range of fundamental skills essential for effective vocal interactions. For the evaluation scheme, we propose several objective evaluation indicators and incorporate an additional LLM-as-a-judge approach to score open-ended questions. Experimental results on 15 mainstream systems reveal significant variability, each exhibiting distinct strengths and weaknesses, and provide valuable insights to guide future research in speech interaction systems.

Hongcheng Liu, Yixuan Hou, Heyang Liu et al.

While Speech Large Language Models (Speech-LLMs) show strong performance in many applications, their robustness is critically under-tested, especially to speech disfluency. Existing evaluations often rely on idealized inputs, overlooking common disfluencies, particularly those associated with conditions like Parkinson's disease. This work investigates whether current Speech-LLMs can maintain performance when interacting with users who have speech impairments. To facilitate this inquiry, we introduce VocalBench-DF, a framework for the systematic evaluation of disfluency across a multi-dimensional taxonomy. Our evaluation of 22 mainstream Speech-LLMs reveals substantial performance degradation, indicating that their real-world readiness is limited. Further analysis identifies phoneme-level processing and long-context modeling as primary bottlenecks responsible for these failures. Strengthening recognition and reasoning capability from components and pipelines can substantially improve robustness. These findings highlight the urgent need for new methods to improve disfluency handling and build truly inclusive Speech-LLMs

Yuhao Wang, Zhiyuan Zhu, Heyang Liu et al.

Multimodal large language models (MLLMs) excel at multimodal perception and understanding, yet their tendency to generate hallucinated or inaccurate responses undermines their trustworthiness. Existing methods have largely overlooked the importance of refusal responses as a means of enhancing MLLMs reliability. To bridge this gap, we present the Information Boundary-aware Learning Framework (InBoL), a novel approach that empowers MLLMs to refuse to answer user queries when encountering insufficient information. To the best of our knowledge, InBoL is the first framework that systematically defines the conditions under which refusal is appropriate for MLLMs using the concept of information boundaries proposed in our paper. This framework introduces a comprehensive data generation pipeline and tailored training strategies to improve the model's ability to deliver appropriate refusal responses. To evaluate the trustworthiness of MLLMs, we further propose a user-centric alignment goal along with corresponding metrics. Experimental results demonstrate a significant improvement in refusal accuracy without noticeably compromising the model's helpfulness, establishing InBoL as a pivotal advancement in building more trustworthy MLLMs.

Heyang Liu, Yu Wang, Yanfeng Wang

End-to-end (E2E) approach is gradually replacing hybrid models for automatic speech recognition (ASR) tasks. However, the optimization of E2E models lacks an intuitive method for handling decoding shifts, especially in scenarios with a large number of domain-specific rare words that hold specific important meanings. Furthermore, the absence of knowledge-intensive speech datasets in academia has been a significant limiting factor, and the commonly used speech corpora exhibit significant disparities with realistic conversation. To address these challenges, we present Medical Interview (MED-IT), a multi-turn consultation speech dataset that contains a substantial number of knowledge-intensive named entities. We also explore methods to enhance the recognition performance of rare words for E2E models. We propose a novel approach, post-decoder biasing, which constructs a transform probability matrix based on the distribution of training transcriptions. This guides the model to prioritize recognizing words in the biasing list. In our experiments, for subsets of rare words appearing in the training speech between 10 and 20 times, and between 1 and 5 times, the proposed method achieves a relative improvement of 9.3% and 5.1%, respectively.