论文解读
不可听水印为何在扩散重生面前失效：DiffErase 的黑盒去除与保真权衡
针对语音/音乐/环境声的神经音频水印，DiffErase 在不查询检测器且不知水印方案的黑盒条件下，通过梅尔频谱的中间噪声扰动与预训练扩散去噪重生实现去除，并在五种水印系统上将检测 TPR 降至近 0 的同时以 t* 控制保真度代价。
扩散模型
论文解读
Chatterbox-Flash: Prior-Calibrated Block Diffusion for Streaming Zero-Shot TTS
语音合成 | 10/10
论文解读
ImmersiveTTS: Environment-Aware Text-to-Speech with Multimodal Diffusion Transformer and Domain-Specific Representation Alignment
语音合成 | 9.3/10
论文解读
SwanVoice: Expressive Long-Form Zero-Shot Speech Synthesis for Both Monologue and Dialogue
语音合成 | 8.9/10
论文解读
Towards Streaming Synchronized Spatial Audio Generation via Autoregressive Diffusion Transformer
自回归模型 | 6.5/10
论文解读
Archon: A Unified Multimodal Model for Holistic Digital Human Generation
语音合成 | 7.5/10
论文解读
Decoding Strategies for Diffusion-Based ASR: A Systematic Evaluation of Confidence-Based Thresholding
语音识别 | 6.8/10
论文解读
Native Audio-Visual Alignment for Generation
音频生成 | 7.8/10
论文解读
Dasheng AudioGen: A Unified Model for Generating Coherent Audio Scenes from Text
音频生成 | 8.6/10
论文解读
DEMON: Diffusion Engine for Musical Orchestrated Noise
音乐生成 | 6.0/10
论文解读
Hidden in Plain Tokens: Simply Robust, Gradient-Free Watermark for Synthetic Audio
语音合成 | 10/10
论文解读
WaveNeXt 2: ConvNeXt-Based Fast Neural Vocoders With Residual Denoising and Sub-Modeling for GAN and Diffusion Models
语音合成 | 8.5/10
论文解读
Diffusion Domain Expansion: Learning to Coordinate Pre-trained Diffusion Models
扩散模型 | 7.4/10
论文解读
FlowLong: Inference-time Long Video Generation via Manifold-constrained Tweedie Matching
视频生成 | 4.9/10
论文解读
Instrumental Text-to-Music Generation with Auxiliary Conditioning Branches
音乐生成 | 8.4/10
论文解读
Raon-OpenTTS: Open Models and Data for Robust Text-to-Speech
语音合成 | 9.5/10
论文解读
A Survey of Advancing Audio Super-Resolution and Bandwidth Extension from Discriminative to Generative Models
音频修复 | 8.1/10
论文解读
Omni-Customizer: End-to-End MultiModal Customization for Joint Audio-Video Generation
音视频 | 7.3/10
论文解读
S2Accompanist: A Semantic-Aware and Structure-Guided Diffusion Model for Music Accompaniment Generation
音乐生成 | 5.6/10
论文解读
SemaVoice: Semantic-Aware Continuous Autoregressive Speech Synthesis
语音合成 | 6.8/10