11 papers
SPRI: SVD-Partitioned Residual Initialization for Data-Constrained MoE Upcycling
Weiqiao Shan, Ruixiang Mao, Yuang Li +10
Mixture-of-Experts (MoE) models enable efficient scaling, but training them from scratch remains prohibitively expensive. MoE upcycling mitigates this cost by converting pretrained…
DirectAudioEdit: Inversion-Free Text-Guided Audio Editing via Diffusion Prediction Contrast
Zhengkun Ge, Xiaoqian Liu, Haoran Zhang +5
Text-guided audio editing aims to modify the language-specified acoustic content while preserving edit-irrelevant source components. Existing training-free methods typically rely o…
On the Emotion Understanding of Synthesized Speech
Yuan Ge, Haishu Zhao, Aokai Hao +10
Emotion is a core paralinguistic feature in voice interaction. It is widely believed that emotion understanding models learn fundamental representations that transfer to synthesize…
StyleBench: Evaluating Speech Language Models on Conversational Speaking Style Control
Haishu Zhao, Aokai Hao, Yuan Ge +3
Speech language models (SLMs) have significantly extended the interactive capability of text-based Large Language Models (LLMs) by incorporating paralinguistic information. For mor…
SageLM: A Multi-aspect and Explainable Large Language Model for Speech Judgement
Yuan Ge, Junxiang Zhang, Xiaoqian Liu +10
Speech-to-Speech (S2S) Large Language Models (LLMs) are foundational to natural human-computer interaction, enabling end-to-end spoken dialogue systems. However, evaluating these m…
M-CIF: Multi-Scale Alignment For CIF-Based Non-Autoregressive ASR
Ruixiang Mao, Xiangnan Ma, Qing Yang +7
The Continuous Integrate-and-Fire (CIF) mechanism provides effective alignment for non-autoregressive (NAR) speech recognition. This mechanism creates a smooth and monotonic mappin…