论文解读
Audio-Visual Intelligence in Large Foundation Models
跨模态 | 8.0/10
生成模型
子标签
下方只统计已核验方向的论文记录，按已核身份去重，身份待核时按页面保留。“直接”只含明确标注该方向的论文；“含子树”同时包含下级方向。点击名称进入论文库继续筛选；数量与上方按标签名称收录的页面数不同。
每日研究速递
语音/音乐/音频论文速递 2026-05-09
共分析 3 篇语音/AI 论文
论文解读
RenCon 2025: Revival of the Expressive Performance Rendering Competition
音乐生成 | 7.0/10
论文解读
Spatial-Magnifier: Spatial upsampling for multichannel speech enhancement
语音增强 | 7.0/10
论文解读
Stage Light is Sequence$^2$: Multi-Light Control via Imitation Learning
音乐信息检索 | 7.5/10
论文解读
MelShield: Robust Mel-Domain Audio Watermarking for Provenance Attribution of AI Generated Synthesized Speech
音频安全 | 7.0/10
论文解读
AUHead: Realistic Emotional Talking Head Generation via Action Units Control
生成模型 | 7.5/10
论文解读
Confident and Adaptive Generative Speech Recognition via Risk Control
语音识别 | 7.5/10
论文解读
DiVeQ: Differentiable Vector Quantization Using the Reparameterization Trick
生成模型 | 8.0/10
会议任务专题
ICLR 2026 - 生成模型
共 2 篇 ICLR 2026 生成模型 方向论文
论文解读
LayerSync: Self-aligning Intermediate Layers
音频生成 | 7.5/10
论文解读
Syncphony: Synchronized Audio-to-Video Generation with Diffusion Transformers
音视频 | 8.0/10
论文解读
Towards Improving Speaker Distance Estimation through Generative Impulse Response Augmentation
声源定位 | 8.5/10
论文解读
A Hidden Semantic Bottleneck in Conditional Embeddings of Diffusion Transformers
生成模型 | 6.5/10
论文解读
AlignSep: Temporally-Aligned Video-Queried Sound Separation with Flow Matching
语音分离 | 8.0/10
论文解读
Automatic Stage Lighting Control: Is it a Rule-Driven Process or Generative Task?
音乐生成 | 8.5/10
论文解读
Flow2GAN: Hybrid Flow Matching and GAN with Multi-Resolution Network for Few-step High-Fidelity Audio Generation
音频生成 | 8.0/10
论文解读
Generative Adversarial Post-Training Mitigates Reward Hacking in Live Human-AI Music Interaction
音乐生成 | 8.0/10
论文解读
LayerSync: Self-aligning Intermediate Layers
生成模型 | 7.5/10
论文解读
Toward Complex-Valued Neural Networks for Waveform Generation
语音合成 | 7.5/10