Mark D. Plumbley

AMemOps: Evaluating Memory Updates in Long Audio Streams

Peiwei Chang, Li Zhang, Mark D. Plumbley +2

Audio memory must recognise recurring sources across acoustic changes while keeping their attributes up to date. Overall accuracy does not reveal how these demands interact. We int…

PhysicsSliders: Learning Physical Controls for Audio Foley Generation

Peiwei Chang, Yongqiang Chen, Yuelan Cheng +6

Physical properties provide intuitive controls for sounds made by interacting objects. The challenge is to make a generative model respond predictably to these controls, both indiv…

cs.AI2026

AVSD-Scenes: A Dataset for Audio-Visual Description of Urban Scenes

Dhanunjaya Varma Devalraju, Arshdeep Singh, Mark D. Plumbley

Natural language descriptions can provide rich semantic representations of audio-visual urban scenes, yet datasets that jointly describe both auditory and visual information remain…

Interpreting hierarchical organisation of speaker embeddings

Yanze Xu, Wenwu Wang, Mark D. Plumbley

Speaker recognition neural networks recognise speaker identities from input utterances by learning latent representations (i.e. speaker embeddings). However, these networks' intern…

Exploring Second-Order Pattern Recognition in Speaker Recognition

Yanze Xu, Wenwu Wang, Mark D. Plumbley

In traditional pattern recognition tasks, neural networks are trained to recognise human-defined patterns (e.g. audio categories) in model inputs (e.g. audio). Meanwhile, some Expl…

FlowSep 2: Self-Supervised Flow Matching for Language-Queried Audio Source Separation

Yi Yuan, Xubo Liu, Haohe Liu +3

Language-queried audio source separation (LASS) aims to extract target sources from audio mixtures according to natural language descriptions, offering a flexible and scalable inte…

Summary of DCASE 2026 Task 5: Audio-Dependent Question Answering

Haolin He, Renhe Sun, Zheqi Dai +16

DCASE~2026 Task~5 introduces Audio-Dependent Question Answering (ADQA), which tests whether large audio-language models answer from the audio rather than from textual priors. An Au…

Efficient Text-to-Audio Generation via Pruning

Arshdeep Singh, Yi Yuan, Yun Chen +2

Diffusion-based text-to-audio generative models such as AudioLDM achieve high perceptual quality and strong semantic consistency; however, their practical deployment is hindered by…

Teaching Speech Enhancement Models to Sing: Domain Adaptation from Speech Enhancement to Singing Voice Separation

Paul A. Bereuter, Mark D. Plumbley, Alois Sontacchi

State-of-the-art speech enhancement models benefit from large-scale labeled datasets, whereas singing voice separation models suffer from limited available training data. To addres…

Explainable AI in Speaker Recognition -- Attention Map Visualisation and Evaluation

Yanze Xu, Mark D. Plumbley, Wenwu Wang

Explaining and understanding the decision-making process of artificial intelligence (AI) systems, particularly those implemented by neural networks, falls within the field of expla…

Explainable AI in Speaker Recognition -- Making Latent Representations Understandable

Yanze Xu, Wenwu Wang, Mark D. Plumbley

Neural networks can be trained to learn task-relevant representations from data. Understanding how these networks make decisions falls within the Explainable AI (XAI) domain. This…

BioDCASE 2026 Challenge Baseline for Cross-Domain Mosquito Species Classification

Yuanbo Hou, Vanja Zdravkovic, Marianne Sinka +5

Mosquito-borne diseases affect more than one billion people each year and cause close to one million deaths. Traditional surveillance methods rely on traps and manual identificatio…

Summary of The Inaugural Music Source Restoration Challenge

Yongyi Zang, Jiarui Hai, Wanying Ge +5

Music Source Restoration (MSR) aims to recover original, unprocessed instrument stems from professionally mixed and degraded audio, requiring the reversal of both production effect…

AudioRAG+: Feedback-driven Retrieval-augmented Audio Generation with Large Audio Language Models

Junqi Zhao, Chenxing Li, Jinzheng Zhao +4

We propose a general feedback-driven retrieval-augmented generation (RAG) approach that leverages Large Audio Language Models (LALMs) to address the missing or imperfect synthesis…

Compressing Quaternion Convolutional Neural Networks for Audio Classification

Arshdeep Singh, Vinayak Abrol, Mark D. Plumbley

Conventional Convolutional Neural Networks (CNNs) in the real domain have been widely used for audio classification. However, their convolution operations process multi-channel inp…

MSRBench: A Benchmarking Dataset for Music Source Restoration

Yongyi Zang, Jiarui Hai, Wanying Ge +5

Music Source Restoration (MSR) extends source separation to realistic settings where signals undergo production effects (equalization, compression, reverb) and real-world degradati…

ISSE: An Instruction-Guided Speech Style Editing Dataset And Benchmark

Yun Chen, Qi Chen, Zheqi Dai +3

Speech style editing refers to modifying the stylistic properties of speech while preserving its linguistic content and speaker identity. However, most existing approaches depend o…

Region-Specific Audio Tagging for Spatial Sound

Jinzheng Zhao, Yong Xu, Haohe Liu +6

Audio tagging aims to label sound events appearing in an audio recording. In this paper, we propose region-specific audio tagging, a new task which labels sound events in a given r…

Measuring Audio's Impact on Correctness: Audio-Contribution-Aware Post-Training of Large Audio Language Models

Haolin He, Xingjian Du, Renhe Sun +16

Large Audio Language Models (LALMs) represent an important frontier in multimodal AI, addressing diverse audio tasks. Recently, post-training of LALMs has received increasing atten…

DreamAudio: Customized Text-to-Audio Generation with Diffusion Models

Yi Yuan, Xubo Liu, Haohe Liu +5

With the development of large-scale diffusion-based and language-modeling-based generative models, impressive progress has been achieved in text-to-audio generation. Despite produc…

Towards Reliable Objective Evaluation Metrics for Generative Singing Voice Separation Models

Paul A. Bereuter, Benjamin Stahl, Mark D. Plumbley +1

Traditional Blind Source Separation Evaluation (BSS-Eval) metrics were originally designed to evaluate linear audio source separation models based on methods such as time-frequency…

AudioTurbo: Fast Text-to-Audio Generation with Rectified Diffusion

Junqi Zhao, Jinzheng Zhao, Haohe Liu +5

Diffusion models have significantly improved the quality and diversity of audio generation but are hindered by slow inference speed. Rectified flow enhances inference speed by lear…