论文解读
不改脸只改声：把保护藏进人耳听不见的掩蔽阴影
语音合成 | 4.3/10
鲁棒性
论文解读
把 2M 维坐标压到一条曲线上：BOGE 如何用物理先验与贝叶斯优化校准水下柔性阵
声源定位 | 5.7/10
论文解读
给大模型加声学缰绳：用 0.6 nats 的似然约束把翻译式幻觉拉回转录
语音识别 | 7.8/10
论文解读
口音不是噪声，是距离：当母语离英语越远，ASR 的潜空间就越把你推开
语音识别 | 6.9/10
论文解读
提示词动了，词错误率为什么不动：一次生产级口述史转录的预注册阴性消融
语音识别 | 7.4/10
论文解读
干净本身成了开关：当语音增强的理想输出反噬自身
语音增强 | 6.4/10
论文解读
别再只调大模型：用 88 行文字把声音的破绽说给 LLM 听
语音伪造检测 | 6.4/10
论文解读
听错一句，机器就敢动手：语音误差如何撬开具身智能的安全锁
语音交互 | 5.1/10
论文解读
先分清状态，再决定能不能说话：Anian 如何把生成关进安全笼子
语音交互 | 5.0/10
论文解读
想找回被替换的原话，先接受水印不能再像哈希那样精确
音频水印 | 6.4/10
论文解读
EM-KalmanNet: Learned Expectation-Maximization for Adaptive Tracking in Partially Known, Block-Wise Time-Varying State-Space Models
声源定位 | 8.8/10
论文解读
Lost in Speech: Trilingual Spoken Hallucination Detection Across Audio and Transcripts
音频理解 | 8.2/10
论文解读
On the Robustness of Audio Deepfake Detection under Audio Watermarking
音频伪造检测 | 7.7/10
论文解读
REDnet: Recursive Encoder and Decoder for Speech Separation under Unknown Number of Speakers and Variable Number of Microphones
语音分离 | 8.2/10
论文解读
Relative Time Intervals Representation for Word-level Timestamping with Masked Training
语音识别 | 8.6/10
论文解读
SonarLLM: A Native Sonar--Optical Multimodal Large Language Model for Underwater Perception
音频理解 | 8.0/10
论文解读
The ISCSLP 2026 Real-World Audio-Visual Speech Enhancement Challenge
音视频语音分离 | 8.8/10
论文解读
Adaptive Hierarchical Representation Alliance for Multimodal Learning
语音情感识别 | 8.6/10
论文解读
AT-ADD: A Benchmark and Challenge for Robust and All-Type Audio Deepfake Detection
音频伪造检测 | 9.6/10
论文解读
AudioNoisePrints: Model-free audio watermarking using spatial correlation in flow matching TTS
音频水印 | 7.9/10