14 papers
What Does LALM Post-training Learn? Rethinking Audio Understanding and Reasoning Capabilities
Hua-Hsuan Chen, Xuanjun Chen, Wei-Chung Lu +5
Post-training often improves the performance of large audio-language models on audio understanding and reasoning benchmarks. However, these improvements may reflect adaptation to s…
Listen, Critique, and Refine: RL-Based Self-Refinement for Instruction-Following Speech Synthesis
Chee-En Yu, Yi-Cheng Lin, Sung-Feng Huang +4
Large Audio Language Models (LALMs) can follow diverse instructions to synthesize speech in specified styles. However, complex instructions that require simultaneous control over p…
HAMMER: Harmonic-Aware Parallel Context Modeling and Discriminator-Free Perceptual Optimization for Speech Enhancement
Shang-Fu Chen, Szu-Wei Fu, Sung-Feng Huang +3
Recent speech enhancement systems combine self-attention and Mamba to capture global interactions and long-range dependencies. Yet these hybrids usually operate as sequence mixers…
RT-SEMamba: Real-Time Speech Enhancement Mamba via Progressive Knowledge Distillation
Rong Chao, Sung-Feng Huang, Moreno La Quatra +4
We present RT-SEMamba, a fully causal speech enhancement (SE) model built upon causal time-frequency Mamba blocks. Unlike Transformer-based architectures that rely on a growing key…
One Model, Many Latencies: Universal Speech Enhancement for Diverse Real-Time Applications
Szu-Wei Fu, Rong Chao, Xuesong Yang +4
Different real-time speech applications impose distinct latency budgets, often requiring separately trained enhancement models for each scenario. In this paper, we propose a one-fo…
Rethinking Training Targets, Architectures and Data Quality for Universal Speech Enhancement
Szu-Wei Fu, Rong Chao, Xuesong Yang +6
Universal Speech Enhancement (USE) aims to restore speech quality under diverse degradation conditions while preserving signal fidelity. Despite recent progress, key challenges in…
How Auditory Knowledge in LLM Backbones Shapes Audio Language Models: A Holistic Evaluation
Ke-Han Lu, Szu-Wei Fu, Chao-Han Huck Yang +13
Large language models (LLMs) have been widely used as knowledge backbones of Large Audio Language Models (LALMs), yet how much auditory knowledge they encode through text-only pre-…
Investigating Safety Vulnerabilities of Large Audio-Language Models Under Speaker Emotional Variations
Bo-Han Feng, Chien-Feng Liu, Yu-Hsuan Li Liang +9
Large audio-language models (LALMs) extend text-based LLMs with auditory understanding, offering new opportunities for multimodal applications. While their perception, reasoning, a…
SAKE: Towards Editing Auditory Attribute Knowledge of Large Audio-Language Models
Chih-Kai Yang, Yen-Ting Piao, Tzu-Wen Hsu +8
Knowledge editing enables targeted updates without retraining, but prior work focuses on textual or visual facts, leaving abstract auditory perceptual knowledge underexplored. We i…
Leveraging Mamba with Full-Face Vision for Audio-Visual Speech Enhancement
Rong Chao, Wenze Ren, You-Jin Li +5
Recent Mamba-based models have shown promise in speech enhancement by efficiently modeling long-range temporal dependencies. However, models like Speech Enhancement Mamba (SEMamba)…
DeSTA2.5-Audio: Toward General-Purpose Large Audio Language Model with Self-Generated Cross-Modal Alignment
Ke-Han Lu, Zhehuai Chen, Szu-Wei Fu +25
We introduce DeSTA2.5-Audio, a general-purpose Large Audio Language Model (LALM) designed for robust auditory perception and instruction-following. Recent LALMs augment Large Langu…
HighRateMOS: Sampling-Rate Aware Modeling for Speech Quality Assessment
Wenze Ren, Yi-Cheng Lin, Wen-Chin Huang +9
Modern speech quality prediction models are trained on audio data resampled to a specific sampling rate. When faced with higher-rate audio at test time, these models can produce bi…
Detecting the Undetectable: Assessing the Efficacy of Current Spoof Detection Methods Against Seamless Speech Edits
Sung-Feng Huang, Heng-Cheng Kuo, Zhehuai Chen +6
Neural speech editing advancements have raised concerns about their misuse in spoofing attacks. Traditional partially edited speech corpora primarily focus on cut-and-paste edits,…
Generative Speech Foundation Model Pretraining for High-Quality Speech Extraction and Restoration
Pin-Jui Ku, Alexander H. Liu, Roman Korostik +3
This paper proposes a generative pretraining foundation model for high-quality speech restoration tasks. By directly operating on complex-valued short-time Fourier transform coeffi…