论文检索

Sangmin Song, Juhwan Choi, JungMin Yun, YoungBin Kim

Large language models (LLMs) have demonstrated remarkable performance in zero-shot dialogue state tracking (DST), reducing the need for task-specific training. However, conventional DST benchmarks primarily focus on structured user-agent conversations, failing to capture the complexities of real-world multi-user interactions. In this study, we assess the robustness of LLMs in multi-user DST while minimizing dataset construction costs. Inspired by recent advances in LLM-based data annotation, we extend an existing DST dataset by generating utterances of a second user based on speech act theory. Our methodology systematically incorporates a second user’s utterances into conversations, enabling a controlled evaluation of LLMs in multi-user settings. Experimental results reveal a significant performance drop compared to single-user DST, highlighting the limitations of current LLMs in extracting and tracking dialogue states amidst multiple speakers. Our findings emphasize the need for future research to enhance LLMs for multi-user DST scenarios, paving the way for more realistic and robust DST models.

Advait Joglekar, Divyanshu Singh, Rooshil Rohit Bhatia, Srinivasan Umesh

Voice Conversion research in recent times has increasingly focused on improving the zero-shot capabilities of existing methods. Despite remarkable advancements, current architectures still tend to struggle in zero-shot cross-lingual settings. They are also often unable to generalize for speakers of unseen languages and accents. In this paper, we adopt a simple yet effective approach that combines discrete speech representations from self-supervised models with a non-autoregressive Diffusion-Transformer based conditional flow matching speech decoder. We show that this architecture allows us to train a voice-conversion model in a purely textless, self-supervised fashion. Our technique works without requiring multiple encoders to disentangle speech features. Our model also manages to excel in zero-shot cross-lingual settings even for unseen languages. We provide our code, model checkpoint and demo samples here: https://github.com/ez-vc/ez-vc

Kshitij Ambilduke, Ben Peters, Sonal Sannigrahi, Anil Keshwani, Tsz Kin Lam, Bruno Martins, Andre Martins, Marcely Zanon Boito

We introduce Spire, a speech-augmented language model (LM) capable of both translating and transcribing speech input from English into 10 other languages as well as translating text input in both language directions. Spire integrates the speech modality into an existing multilingual LM via speech discretization and continued pre-training using only 42.5 K hours of speech. In particular, we adopt the pretraining framework of multilingual LMs and treat discretized speech input as an additional translation language. This approach not only equips the model with speech capabilities, but also preserves its strong text-based performance. We achieve this using significantly less data than existing speech LMs, demonstrating that discretized speech input integration as an additional language is feasible during LM adaptation. We make our code and models available to the community.

Hend ElGhazaly, Bahman Mirheidari, Heidi Christensen, Nafise Sadat Moosavi

Modern Automatic Speech Recognition (ASR) systems are increasingly deployed in high-stakes settings, including clinical interviews, public services, and educational tools, where equitable performance across speaker groups is essential. While pre-trained speech models like Whisper achieve strong overall accuracy, they often exhibit inconsistent group-level performance that varies across domains. These disparities are not fixed properties of the model, but emerge from the interaction between model, data, and task—posing challenges for fairness interventions designed in-domain.We frame fairness in ASR as a generalisation problem. We fine-tune a Whisper model on the Fair-Speech corpus using four strategies: basic fine-tuning, demographic rebalancing, gender-swapped data augmentation, and a novel contrastive learning objective that encourages gender-invariant representations. We evaluate performance across multiple aspects of fairness and utility, both in-domain and on three out-of-domain test sets: LibriSpeech, EdAcc, and CognoSpeak.Our findings show that the method with the best in-domain fairness performed worst out-of-domain, illustrating that fairness gains do not always generalise. Demographic balancing generalises more consistently, while our contrastive method offers a practical alternative: it achieves stable, cross-domain fairness improvements without requiring changes to the training data distribution, and with minimal accuracy trade-offs.

Pengcheng Li, Botao Zhao, Zuheng Kang, Junqing Peng, Xiaoyang Qu, Yayun He, Jianzong Wang

Although large audio-language models (LALMs) have demonstrated remarkable capabilities in audio perception, their performance in affective computing scenarios, particularly in emotion recognition, reasoning, and subtle sentiment differentiation, remains suboptimal. Recent advances in reinforcement learning (RL) have shown promise in improving LALMs’ reasoning abilities. However, two critical challenges hinder the direct application of RL techniques to speech emotion recognition (SER) tasks: (1) convergence instability caused by ambiguous emotional boundaries and (2) limited reasoning ability when using relatively small models (e.g., 7B-parameter architectures). To address these challenges, we propose EMO-RL, a novel framework incorporating reinforcement learning with two key innovations: Emotion Similarity-Weighted Reward (ESWR) and Explicit Structured Reasoning (ESR). Built upon pretrained LALMs, our method employs group-relative policy optimization with emotion constraints. Comprehensive experiments demonstrate that our EMO-RL training strategies can significantly enhance the emotional reasoning capabilities of LALMs, achieving state-of-the-art performance on the MELD and IEMOCAP datasets, and cross-dataset experiments prove the strong superiority of generalization.

Siyu Tian, Kaijie Mo, Yupei Wang, Renfen Hu

Multi-turn dialogue is a key paradigm for interaction between users and Large Language Models (LLMs). However, existing evaluation benchmarks fail to capture users’ evolving needs and how their diverse conversation styles affect the dialogue flow. To address these limitations, we propose CMT-Eval, the first dedicated dataset for fine-grained evaluation of Chinese multi-turn dialogue systems. Built upon a linguistic theory-driven Speech Act Framework, diverse user personas, and varied conversational challenges, CMT-Eval comprises 596 high-quality dialogues with 4,431 turns, simulating realistic, multifaceted, and challenging conversations. Experiments reveal that models struggle with specific speech acts, user personas, and complex scenarios, highlighting the effectiveness of CMT-Eval in assessing LLMs’ multi-turn dialogue capabilities and providing valuable insights for their enhancement. The dataset, code, and prompts are available at https://github.com/hejaida/CMT-Eval.

Rao Ma, Mengjie Qian, Vyas Raina, Mark Gales, Kate Knill

The combination of pre-trained speech encoders with large language models has enabled the development of speech LLMs that can handle a wide range of spoken language processing tasks. While these models are powerful and flexible, this very flexibility may make them more vulnerable to adversarial attacks. To examine the extent of this problem, in this work we investigate universal acoustic adversarial attacks on speech LLMs. Here a fixed, universal, adversarial audio segment is prepended to the original input audio. We initially investigate attacks that cause the model to either produce no output or to perform a modified task overriding the original prompt. We then extend the nature of the attack to be selective so that it activates only when specific input attributes, such as a speaker gender or spoken language, are present. Inputs without the targeted attribute should be unaffected, allowing fine-grained control over the model outputs. Our findings reveal critical vulnerabilities in Qwen2-Audio and Granite-Speech and suggest that similar speech LLMs may be susceptible to universal adversarial attacks. This highlights the need for more robust training strategies and improved resistance to adversarial attacks.

Rao Ma, Tongzhou Chen, Kartik Audhkhasi, Bhuvana Ramabhadran

Recently, large-scale pre-trained speech encoders and Large Language Models (LLMs) have been released, which show state-of-the-art performance on a range of spoken language processing tasks, including Automatic Speech Recognition (ASR). To effectively combine both models for better performance, continuous speech prompts and ASR error correction have been adopted. However, these methods are prone to suboptimal performance or are inflexible. In this paper, we propose a new paradigm, LegoSLM, that bridges speech encoders and LLMs using the ASR posterior matrices. The speech encoder is trained to generate Connectionist Temporal Classification (CTC) posteriors over the LLM vocabulary, which are used to reconstruct pseudo-audio embeddings by computing a weighted sum of the LLM input embeddings. These embeddings are concatenated with text embeddings in the LLM input space. Using the well-performing USM and Gemma models as an example, we demonstrate that our proposed LegoSLM method yields good performance on both ASR and speech translation tasks. By connecting USM with Gemma models, we can get an average of 49% WER reduction (WERR) over the USM-CTC baseline on 8 MLS testsets. The trained model also exhibits modularity in a range of settings – after fine-tuning the Gemma model weights, the speech encoder can be switched and combined with the LLM in a zero-shot fashion. Additionally, we propose to control the decode-time influence of the USM and LLM using a softmax temperature, which shows effectiveness in domain adaptation.

Ruiqi Yan, Xiquan Li, Wenxi Chen, Zhikang Niu, Chen Yang, Ziyang Ma, Kai Yu, Xie Chen

Recent advances in large language models (LLMs) have driven significant progress in end-to-end spoken dialogue models (SDMs). In contrast to text-based LLMs, the evaluation framework for SDMs should encompass both cognitive dimensions (e.g., logical reasoning, knowledge) and speech-related aspects (e.g., paralinguistic cues, audio quality). However, there is still a lack of comprehensive evaluations for SDMs in speech-to-speech (S2S) scenarios. To address this gap, we propose **URO-Bench**, an extensive benchmark for SDMs. Notably, URO-Bench is the first S2S benchmark that covers evaluations about multilingualism, multi-round dialogues, and paralinguistics. Our benchmark is divided into two difficulty levels: basic track and pro track, each comprising 20 test sets, evaluating the spoken dialogue model’s abilities in **U**nderstanding, **R**easoning, and **O**ral conversation. Evaluations on our proposed benchmark reveal that current open-source SDMs perform rather well in daily QA tasks, but lag behind their backbone LLMs in terms of instruction-following ability and also suffer from catastrophic forgetting. Their performance in advanced evaluations of paralinguistic information and audio understanding remains subpar, highlighting the need for further research in this direction. We hope that URO-Bench can facilitate the development of spoken dialogue models by providing a multifaceted evaluation of existing models and helping to track progress in this area.

Valentin Barriere, Nahuel Gomez, Léo Hemamou, Sofia Callejas, Brian Ravenet

Aiming towards improving current computational models of humor detection, we propose a new multimodal dataset of stand-up comedies, in seven languages: English, French, Spanish, Italian, Portuguese, Hungarian and Czech. Our dataset of more than 330 hours %, is at the time of writing the biggest available for this type of task, and the most diverse. The whole dataset is automatically annotated in laughter (from the audience), and the subpart left for model validation is manually annotated.% Contrary to contemporary approaches, we do not frame the task of humor detection as a binary sequence classification, but as word-level sequence labeling, in order to take into account all the context of the sequence and to capture the continuous joke tagging mechanism typically occurring in natural conversations. As par with unimodal baselines results, we propose a method for e propose a method to enhance the automatic laughter detection based on Audio Speech Recognition errors. Our code and data are available online: https://tinyurl.com/EMNLPHumourStandUpAnonym

Kaiying Kevin Lin, Hsi-Yu Chen, Haopeng Zhang

While large language models (LLMs) have demonstrated impressive performance across a wide range of natural language processing (NLP) tasks in high-resource languages, their capabilities in low-resource and minority languages remain significantly underexplored. Formosan languages—a subgroup of Austronesian languages spoken in Taiwan—are both linguistically rich and endangered, largely due to the sociolinguistic dominance of Mandarin. In this work, we introduce FormosanBench, the first benchmark for evaluating LLMs on low-resource Austronesian languages. It covers three endangered Formosan languages: Atayal, Amis, and Paiwan, across three core NLP tasks: machine translation, automatic speech recognition (ASR), and text summarization. We assess model performance in zero-shot, 10-shot, and fine-tuned settings using FormosanBench. Our results reveal a substantial performance gap between high-resource and Formosan languages. Existing LLMs consistently underperform across all tasks, with 10-shot learning and fine-tuning offering only limited improvements. These findings underscore the urgent need for more inclusive NLP technologies that can effectively support endangered and underrepresented languages. We release our datasets and code to facilitate future research in this direction :https://anonymous.4open.science/r/FormosanBench-DB43/

Huu Tuong Tu, Huan Vu, Cuong Tien Nguyen, Dien Hy Ngo, Nguyen Thi Thu Trang

Traditional voice conversion (VC) methods typically attempt to separate speaker identity and linguistic information into distinct representations, which are then combined to reconstruct the audio. However, effectively disentangling these factors remains challenging, often leading to information loss during training. In this paper, we propose a new approach that leverages synthetic speech data generated by a high-quality, pretrained multispeaker text-to-speech (TTS) model. Specifically, synthetic data pairs that share the same linguistic content but differ in speaker identity are used as input-output pairs to train the voice conversion model. This enables the model to learn a direct mapping between source and target voices, effectively capturing speaker-specific characteristics while preserving linguistic content. Additionally, we introduce a flexible training strategy for any-to-any voice conversion that generalizes well to unseen speakers and new languages, enhancing adaptability and performance in zero-shot scenarios. Our experiments show that our proposed method achieves a 16.35% relative reduction in word error rate and a 5.91% improvement in speaker cosine similarity, outperforming several state-of-the-art methods. Voice conversion samples can be accessed at: https://oovc-emnlp-2025.github.io/

Jackson Trager, Francielle Vargas, Diego Alves, Matteo Guida, Mikel K. Ngueajio, Ameeta Agrawal, Yalda Daryani, Farzan Karimi Malekabadi, Flor Miriam Plaza-del-Arco

Ensuring the moral reasoning capabilities of Large Language Models (LLMs) is a growing concern as these systems are used in socially sensitive tasks. Nevertheless, current evaluation benchmarks present two major shortcomings: a lack of annotations that justify moral classifications, which limits transparency and interpretability; and a predominant focus on English, which constrains the assessment of moral reasoning across diverse cultural settings. In this paper, we introduce MFTCXplain, a multilingual benchmark dataset for evaluating the moral reasoning of LLMs via multi-hop hate speech explanations using the Moral Foundations Theory. MFTCXplain comprises 3,000 tweets across Portuguese, Italian, Persian, and English, annotated with binary hate speech labels, moral categories, and text span-level rationales. Our results show a misalignment between LLM outputs and human annotations in moral reasoning tasks. While LLMs perform well in hate speech detection (F1 up to 0.836), their ability to predict moral sentiments is notably weak (F1 < 0.35). Furthermore, rationale alignment remains limited mainly in underrepresented languages. Our findings show the limited capacity of current LLMs to internalize and reflect human moral reasoning.

Tadesse Destaw Belay, Israel Abebe Azime, Ibrahim Said Ahmad, David Ifeoluwa Adelani, Idris Abdulmumin, Abinew Ali Ayele, Shamsuddeen Hassan Muhammad, Seid Muhie Yimam

Language models built from various sources are the foundation of today’s NLP progress. However, for many low-resource languages, the diversity of domains is often limited, more biased to a religious domain, which impacts their performance when evaluated on distant and rapidly evolving domains such as social media. Domain adaptive pre-training (DAPT) and task-adaptive pre-training (TAPT) are popular techniques to reduce this bias through continual pre-training for BERT-based models, but they have not been explored for African multilingual encoders. In this paper, we explore DAPT and TAPT continual pre-training approaches for African languages social media domain. We introduce AfriSocial, a large-scale social media and news domain corpus for continual pre-training on several African languages. Leveraging AfriSocial, we show that DAPT consistently improves performance (from 1% to 30% F1 score) on three subjective tasks: sentiment analysis, multi-label emotion, and hate speech classification, covering 19 languages. Similarly, leveraging TAPT on the data from one task enhances performance on other related tasks. For example, training with unlabeled sentiment data (source) for a fine-grained emotion classification task (target) improves the baseline results by an F1 score ranging from 0.55% to 15.11%. Combining these two methods (i.e. DAPT + TAPT) further improves the overall performance. The data and model resources are available at HuggingFace.

Yang Han, Jacqueline CK Lam, Victor OK Li, Lawrence YL Cheung

Alzheimer’s Disease (AD), the 7th leading cause of death globally, demands scalable methods for early detection. While speech-based diagnostics offer promise, existing approaches struggle with temporal-spatial (T-S) challenges in capturing subtle linguistic shifts across different disease stages (temporal) and in adapting to cross-linguistic variability (spatial). This study introduces a novel Large Language Model (LLM)-driven T-S fusion framework that integrates multilingual LLMs, contrastive learning, and interpretable marker discovery to revolutionize Late Onset AD (LOAD) detection. Our key innovations include: (1) T-S Data Imputation: Leveraging LLMs to generate synthetic speech transcripts across different LOAD stages (NC, Normal Control; eMCI, early Mild Cognitive Impairment; lMCI, late Mild Cognitive Impairment; AD) and languages (Chinese, English, Spanish), addressing data scarcity while preserving clinical relevance (expert validation: 86% agreement with LLM-generated labels). (2) T-S Transformer with Contrastive Learning: A multilingual model that disentangles stage-specific (temporal) and language-specific (spatial) patterns, achieving a notable improvement of 10.9–24.7% in F1-score over existing baselines. (3) Cross-Linguistic Marker Discovery: Identifying language-agnostic markers and language-specific patterns to enhance interpretability for clinical adoption. By unifying temporal LOAD stages and spatial diversity, our framework achieves state-of-the-art performance in early LOAD detection while enabling cross-linguistic diagnostics. This study bridges NLP and clinical neuroscience, demonstrating LLMs’ potential to amplify limited biomedical data and advance equitable healthcare AI.

Nicola Arici, Luca Putelli, Ejdis Gjinika, Ivan Serina, Alfonso Emilio Gerevini

Figurative language and figures of speech, such as metaphors and hyperboles, are used every day in written and oral communication among human beings. Nonetheless, this imaginative use of words in a non literal way requires a solid understanding of semantics and a deep real-world knowledge. In the longstanding debate about whether Neural Language Models (NLMs) really have a full understanding of text, analysing how they can recognise figurative language can provide some intuition of their functioning, their capabilities and their limits. Therefore, in this paper, we exploit probing tasks to study how several NLMs of different sizes recognise four different figures of speech: hyperboles, metaphors, oxymorons and pleonasms. We analyse whether this information is learned and how it is acquired during the training of the model, describing its learning trajectory. Moreover, we analyse which layers have a better comprehension of figurative language and the influence of pre-training data. Datasets and code are available at https://github.com/nicolarici/learning-trajectories.

Siskos Dimitrios, Stavros Papadopoulos, Pablo Peso Parada, Jisi Zhang, Karthikeyan Saravanan, Anastasios Drosou

This work investigates retrieval augmented generation as an efficient strategy for automatic context discovery in context-aware Automatic Speech Recognition (ASR) system, in order to improve transcription accuracy in the presence of rare or out-of-vocabulary terms. However, identifying the right context automatically remains an open challenge. This work proposes an efficient embedding-based retrieval approach for automatic context discovery in ASR. To contextualize its effectiveness, two alternatives based on large language models (LLMs) are also evaluated: (1) large language model (LLM)-based context generation via prompting, and (2) post-recognition transcript correction using LLMs. Experiments on the TED-LIUMv3, Earnings21 and SPGISpeech demonstrate that the proposed approach reduces WER by up to 17% (percentage difference) relative to using no-context, while the oracle context results in a reduction of up to 24.1%.

Matej Martinc, Thi Hong Hanh Tran, Senja Pollak, Boshko Koloski

Keyword extraction involves identifying the most descriptive words in a document, allowing automatic categorisation and summarisation of large quantities of diverse textual data. Relying on the insight that real-world keyword detection often requires handling of diverse content, we propose a novel supervised keyword extraction approach based on the mixture of experts (MoE) technique. MoE uses a learnable routing sub-network to direct information to specialised experts, allowing them to specialise in distinct regions of the input space. SEKE, a mixture of Specialised Experts for supervised Keyword Extraction, uses DeBERTa as the backbone model and builds on the MoE framework, where experts attend to each token, by integrating it with a bidirectional Long short-term memory (BiLSTM) network, to allow successful extraction even on smaller corpora, where specialisation is harder due to lack of training data. The MoE framework also provides an insight into inner workings of individual experts, enhancing the explainability of the approach. We benchmark SEKE on multiple English datasets, achieving state-of-the-art performance compared to strong supervised and unsupervised baselines. Our analysis reveals that depending on data size and type, experts specialise in distinct syntactic and semantic components, such as punctuation, stopwords, parts-of-speech, or named entities. Code is available at https://github.com/matejMartinc/SEKE_keyword_extraction.

Robert Litschko, Verena Blaschke, Diana Burkhardt, Barbara Plank, Diego Frassinelli

Dialects exhibit a substantial degree of variation due to the lack of a standard orthography. At the same time, the ability of Large Language Models (LLMs) to process dialects remains largely understudied. To address this gap, we use Bavarian as a case study and investigate the lexical dialect understanding capability of LLMs by examining how well they recognize and translate dialectal terms across different parts-of-speech. To this end, we introduce DiaLemma, a novel annotation framework for creating dialect variation dictionaries from monolingual data only, and use it to compile a ground truth dataset consisting of 100K human-annotated German-Bavarian word pairs. We evaluate how well nine state-of-the-art LLMs can judge Bavarian terms as dialect translations, inflected variants, or unrelated forms of a given German lemma. Our results show that LLMs perform best on nouns and lexically similar word pairs, and struggle most in distinguishing between direct translations and inflected variants. Interestingly, providing additional context in the form of example usages improves the translation performance, but reduces their ability to recognize dialect variants. This study highlights the limitations of LLMs in dealing with orthographic dialect variation and emphasizes the need for future work on adapting LLMs to dialects.

Qiongqiong Wang, Hardik Bhupendra Sailor, Tianchi Liu, Wenyu Zhang, Muhammad Huzaifah, Nattadaporn Lertcheva, Shuo Sun, Nancy F. Chen, Jinyang Wu, AiTi Aw

Recent speech-LLMs have shown impressive performance in tasks like transcription and translation, yet they remain limited in understanding the paralinguistic aspects of speech crucial for social and emotional intelligence. We propose CP-Bench, a benchmark for evaluating speech-LLMs on contextual paralinguistic reasoning the integration of verbal content with non-verbal cues like emotion and prosody. The benchmark includes two curated question answering (QA) datasets requiring both linguistic and empathetic understanding. We evaluate state-of-the-art speech-LLMs from both open and closed-source models and perform a comprehensive analysis across different question types. The top two models were further analyzed under temperature tuning to understand its effect on this task. Our benchmark reveals a key gap in existing evaluations and offers insights into building more context-aware and emotionally intelligent speech-capable LLMs.