论文解读
绕一圈回不来：用圆形缓冲器讲清莫比乌斯带的单值性与缠绕数
论文把循环移位读成离散圆上的线丛转移映射，用绕行一周的单值性与回到恒等的缠绕数统一解释莫比乌斯波导的基频减半与延迟环混沌振荡器的次谐波，代价是只做解析构造与谱推演而无可运行的定量声学实验对照。
理论分析
论文解读
实增益掩蔽的天花板是投影，离开直线却赢不回平方误差
论文刻画单通道时频实增益掩蔽的正交投影上限，并用一个高斯混合后验均值估计器研究先验、读出与相位对称性：长窗留出设置下估计器距实测类上界 11.44 dB，校准分析中的约 70% 是分贝缺口之比，不是全部缺口或误差能量都由后验方差解释，更不是所有生成分离器的性能上限。
论文解读
把拉格从玄学拉回有限状态：当重建变成约束 MAP 的精确动态规划
音乐理解 | 6.8/10
论文解读
方差不该越过底线：用单峰约束重画 MOS 的容许区间
语音质量评估 | 7.4/10
论文解读
当和声不再只谈音高：用搬运代价度量音色的几何
音乐理解 | 5.5/10
论文解读
Arbitrary Polygon Oscillator: Generalizing Polygonal Synthesis to Arbitrary Shapes, Morphing, and Three-Dimensional Polyhedra
音乐生成 | 8.7/10
论文解读
From local kernels to global form: modeling the emergence of musical content
音乐理解 | 8.2/10
论文解读
A Computationally Efficient Likelihood Approximation for Target Tracking in Time-Varying Multipath Channels
声源定位 | 9.8/10
论文解读
A Regularized Block Diagonal RLS Algorithm for Acoustic Echo Cancellation
回声消除 | 7.2/10
论文解读
LipsAM: Lipschitz-continuous Neural Networks for Convergent Plug-and-Play Audio Signal Recovery
音频修复 | 10.0/10
论文解读
Does Mapping Non-Maximal Probabilities to GMM Components Matter for S-JEPA Encoder Representations?
音频理解 | 5.6/10
论文解读
Why GPT-Style Models Do Not Directly Transfer to Symbolic Music: Compression in the Wrong Coordinate System
音乐生成 | 6.7/10
论文解读
AudioTQ: A Data-Oblivious 6-Bit CPU Audio Codec via Randomized Hadamard Rotation and Lloyd-Max Quantization
音频编码 | 3.5/10
论文解读
Evidence of Absence: Cross-Modal Abductive Risk Perception to Sustain World Models When Vision Fails
音频事件检测 | 6.6/10
论文解读
Impulse Response Estimation via Laguerre-Fourier Expansion
音频理解 | 6.8/10
论文解读
The MPB Corpus: A Dataset of Melody, Rhythm, Harmony, and Melody-Harmony Relationships in Brazilian Popular Music
音乐理解 | 6.8/10
论文解读
Why Performance Metrics Overpromise in Auditory Attention Decoding: an Information-Theoretic Reappraisal
语音交互 | 6.9/10
论文解读
In Defense of Using Worst-case Privacy Disclosure as Privacy Evaluation Metric of Voice Anonymization
说话人验证 | 7.1/10
论文解读
Modeling and Interpreting Correlations, Null Distributions and Significance Levels in Neural Tracking of Natural Stimuli
音频理解 | 8.0/10
论文解读
The Voiceprint Fallacy: Why Voices Are Not Unique Biometric Imprints
说话人验证 | 6.0/10