论文解读
Leveraging Text-to-Speech and Voice Conversion as Data Augmentation for Alzheimer's Disease Detection from Spontaneous Speech
语音生物标志物 | 7.0/10
语音生物标志物
论文解读
Low-Resource Speech-Based Early Alzheimers Detection via Cross-Lingual and Few-Shot Transfer Learning
语音生物标志物 | 7.5/10
论文解读
Mixture of Experts for Recognizing Depression from Interview and Reading Tasks
语音生物标志物 | 6.0/10
论文解读
Modeling Inter-Segment Relationships in Speech for Dementia Detection with Audio Spectrogram Transformers and Graph Attention Networks
语音生物标志物 | 7.0/10
论文解读
Multi-View Hierarchical Hypergraph Neural Network for Automatic Stuttering Detection
语音生物标志物 | 7.5/10
论文解读
Multimodal LLMs as Expert Speech Annotators: Acoustic Macro-Descriptors for Parkinson's Detection
语音生物标志物 | 6.5/10
论文解读
Obstructive Sleep Apnea Endotype Prediction During Wakefulness Using Voice Biomarkers
语音生物标志物 | 6.5/10
论文解读
Optimizing Domain-Adaptive Self-Supervised Learning for Clinical Voice-Based Disease Classification
语音生物标志物 | 7.0/10
论文解读
Probing Whisper for Dysarthric Speech in Detection and Assessment
语音生物标志物 | 6.5/10
论文解读
Reliable AI via Age-Balanced Validation: Fair Model Selection for Parkinson’s Detection from Voice
语音生物标志物 | 7.5/10
论文解读
When Children Talk and Machines Listen: Toward an Interpretable Speech-Based Screener for Dutch Developmental Language Disorder
语音生物标志物 | 7.0/10
论文解读
Comparison of sEMG Encoding Accuracy Across Speech Modes Using Articulatory and Phoneme Features
语音生物标志物 | 8.0/10
论文解读
Time vs. Layer: Locating Predictive Cues for Dysarthric Speech Descriptors in wav2vec 2.0
语音生物标志物 | 7.0/10
论文解读
SAND: The Challenge on Speech Analysis for Neurodegenerative Disease Assessment
1. **解决的问题**：针对神经退行性疾病（特别是肌萎缩侧索硬化症ALS）的早期诊断和监测，缺乏大规模、有临床标注的语音数据集，以及标准化的算法评估框架。 2. **方法核心**：构建并发布了名为SAND的挑战赛，其核心是提供一个扩展的、包含纵向数据的ALS患者与健康对照语音数据集（VOC-A
论文解读
Comparison of sEMG Encoding Accuracy Across Speech Modes Using Articulatory and Phoneme Features
这篇论文旨在为无声言语接口（SSI）选择更优的中间表示目标。研究系统比较了发音特征（SPARC）和传统的音素独热编码，在预测表面肌电（sEMG）信号包络上的表现。核心发现是：1）在出声、默语和次发声三种模式下，SPARC特征的编码准确性均显著优于音素特征；2）出声和默语模式的编码性能相当，次发声模式
论文解读
The Acoustic Camouflage Phenomenon: Re-evaluating Speech Features for Financial Risk Prediction
本研究探讨了在企业财报电话会议中，副语言声学特征（音高、抖动、停顿等）对预测灾难性股价下跌的效用。作者基于MAEC数据集，提取了两种模态的特征：文本端使用FinBERT计算脚本化开场白与即兴Q&A之间的情感极性差异（Sentiment Delta），音频端提取临床语音压力标记的方差特征（音高方差、抖
论文解读
Who is Speaking or Who is Depressed? A Controlled Study of Speaker Leakage in Speech-Based Depression Detection
这篇论文的核心贡献在于系统性地揭示并量化了语音抑郁症检测模型中普遍存在的“说话人身份泄露”问题。作者指出，当前许多报告高准确率的模型，其性能可能严重依赖于对说话人身份（声纹）的记忆，而非对抑郁相关声学