18 papers
Beyond Uniform Subspaces: Spectrum-Aware and Depth-Adaptive Fusion for Multi-Task Model Merging
Ruxi Gu, Zilei Wang, Wei Wang
Model merging aims to consolidate multiple task-specific models without access to extra training process. However, existing subspace-based methods largely rely on a uniform treatme…
Luna-TTS Family Technical Report
Feng Yin, Shuai Shi, Junjie Zheng +19
Modern text-to-speech (TTS) is dominated by autoregressive (AR) codec language models, whose left-to-right decoding brings latency that grows with utterance length, error accumulat…
On the Distillation Loss Functions of Speech VAE for Unified Reconstruction, Understanding, and Generation
Changhao Cheng, Wei Wang, Wangyou Zhang +4
Continuous speech representations based on Variational Autoencoders (VAEs) have emerged as a promising alternative to traditional spectrogram or discrete token based features for s…
LegoDiffusion: Micro-Serving Text-to-Image Diffusion Workflows
Lingyun Yang, Suyi Li, Tianyu Feng +10
Text-to-image generation executes a diffusion workflow comprising multiple models centered on a base diffusion model. Existing serving systems treat each workflow as an opaque mono…
Exploring Motion-Language Alignment for Text-driven Motion Generation
Ruxi Gu, Zilei Wang, Wei Wang
Text-driven human motion generation aims to synthesize realistic motion sequences that follow textual descriptions. Despite recent advances, accurately aligning motion dynamics wit…
DSFlow: Dual Supervision and Step-Aware Architecture for One-Step Flow Matching Speech Synthesis
Bin Lin, Peng Yang, Chao Yan +5
Flow-matching models have enabled high-quality text-to-speech synthesis, but their iterative sampling process during inference incurs substantial computational cost. Although disti…
Representation-Regularized Convolutional Audio Transformer for Audio Understanding
Bing Han, Chushu Zhou, Yifan Yang +4
Bootstrap-based Self-Supervised Learning (SSL) has achieved remarkable progress in audio understanding. However, existing methods typically operate at a single level of granularity…
SLM-SS: Speech Language Model for Generative Speech Separation
Tianhua Li, Chenda Li, Wei Wang +4
Speech separation (SS) has advanced significantly with neural network-based methods, showing improved performance on signal-level metrics. However, these methods often struggle to…
UrgentMOS: Unified Multi-Metric and Preference Learning for Robust Speech Quality Assessment
Wei Wang, Wangyou Zhang, Chenda Li +12
Automatic speech quality assessment has become increasingly important as modern speech generation systems continue to advance, while human listening tests remain costly, time-consu…
ICASSP 2026 URGENT Speech Enhancement Challenge
Chenda Li, Wei Wang, Marvin Sach +8
The ICASSP 2026 URGENT Challenge advances the series by focusing on universal speech enhancement (SE) systems that handle diverse distortions, domains, and input conditions. This o…
Inferix: A Block-Diffusion based Next-Generation Inference Engine for World Simulation
Inferix Team, Tianyu Feng, Yizeng Han +13
World models serve as core simulators for fields such as agentic AI, embodied AI, and gaming, capable of generating long, physically realistic, and interactive high-quality videos.…
MeanSE: Efficient Generative Speech Enhancement with Mean Flows
Jiahe Wang, Hongyu Wang, Wei Wang +5
Speech enhancement (SE) improves degraded speech's quality, with generative models like flow matching gaining attention for their outstanding perceptual quality. However, the flow-…
P.808 Multilingual Speech Enhancement Testing: Approach and Results of URGENT 2025 Challenge
Marvin Sach, Yihui Fu, Kohei Saijo +9
In speech quality estimation for speech enhancement (SE) systems, subjective listening tests so far are considered as the gold standard. This should be even more true considering t…
URGENT-PK: Perceptually-Aligned Ranking Model Designed for Speech Enhancement Competition
Jiahe Wang, Chenda Li, Wei Wang +11
The Mean Opinion Score (MOS) is fundamental to speech quality assessment. However, its acquisition requires significant human annotation. Although deep neural network approaches, s…
Lessons Learned from the URGENT 2024 Speech Enhancement Challenge
Wangyou Zhang, Kohei Saijo, Samuele Cornell +10
The URGENT 2024 Challenge aims to foster speech enhancement (SE) techniques with great universality, robustness, and generalizability, featuring a broader task definition, large-sc…
Interspeech 2025 URGENT Speech Enhancement Challenge
Kohei Saijo, Wangyou Zhang, Samuele Cornell +9
There has been a growing effort to develop universal speech enhancement (SE) to handle inputs with various speech distortions and recording conditions. The URGENT Challenge series…
Improving Speech Enhancement with Multi-Metric Supervision from Learned Quality Assessment
Wei Wang, Wangyou Zhang, Chenda Li +3
Speech quality assessment (SQA) aims to predict the perceived quality of speech signals under a wide range of distortions. It is inherently connected to speech enhancement (SE), wh…
LawDNet: Enhanced Audio-Driven Lip Synthesis via Local Affine Warping Deformation
Deng Junli, Luo Yihao, Yang Xueting +4
In the domain of photorealistic avatar generation, the fidelity of audio-driven lip motion synthesis is essential for realistic virtual interactions. Existing methods face two key…