论文解读
SALMONN-2: Advancing General-Purpose Hearing Abilities with Self-Supervised Representations
音频理解 | 9.4/10
音频理解
论文解读
Should Missing Modalities Always Be Necessary to Repair for Multi-modal Sentiment Analysis?
多模态模型 | 7.0/10
论文解读
SSTMark: Robust Training-Free Semantic-Level Speech Watermarking
音频水印 | 6.5/10
论文解读
Team RAS in 11th ABAW Competition: Multimodal Ambivalence Recognition Approach
语音情感识别 | 5.3/10
论文解读
The tttAI System for the TSA-ASR Task of the SmartGlasses Challenge 2026
说话人日志 | 6.5/10
论文解读
Time-Frequency Consistency Learning for Robust Speech Deepfake Detection
语音伪造检测 | 7.9/10
论文解读
When to Use Extra Context: Evidence-Grounded Terminology Adaptation for Simultaneous Speech Translation
语音翻译 | 6.7/10
论文解读
A Geometry-Limited Identification Floor and Its Consequences for Voice-Clone Attribution in Professional Voice Actors
说话人验证 | 8.8/10
论文解读
A Study of Parallelizable Alternatives to Dynamic Time Warping for Aligning Long Sequences
基准测试 | 8.1/10
论文解读
AnovaX: A Local, Multi-Agent Voice Assistant with LLM Planning, Typed Executors, and Adaptive Recovery
语音交互 | 4.8/10
论文解读
AuEmoChat: Authentic Emotion Understanding and Rendering for Conversational Speech Synthesis
语音合成 | 6.9/10
论文解读
AV-JEPA: Extending LeJEPA to Audio-Visual Self-Supervised Learning
音视频理解 | 5.7/10
论文解读
Constrained Hebbian Learning Supports Efficient Representational Allocation under Structural Constraints
音视频理解 | 6.7/10
论文解读
Controlling Implicit Shortcut Reliance in L2 Spoken English Auto-markers
语音质量评估 | 7.5/10
论文解读
Data-driven Video Codec with Implicit Neural Representations
音频编码 | 5.3/10
论文解读
Estimating the Reliability of Dynamic Time Warping Alignments Using Circumstantial Evidence
音乐理解 | 7.6/10
论文解读
Natural Backdoor Attacks on Speech Recognition Models
语音识别 | 3.5/10
论文解读
Proof-Carrying Multimodal Timelines: Finite-Trace Modal Certificates for Video-Audio Consistency
基准测试 | 8.6/10
论文解读
Segmental DTW: A Parallelizable Alternative to Dynamic Time Warping
音频检索 | 7.0/10
论文解读
SpeechGuard: Online Defense against Backdoor Attacks on Speech Recognition Models
语音识别 | 6.0/10