论文解读
Variational Low-Rank Adaptation for Personalized Impaired Speech Recognition
语音识别 | 7.5/10
语音识别
论文解读
Voting-Based Pitch Estimation with Temporal and Frequential Alignment and Correlation Aware Selection
语音识别 | 8.0/10
论文解读
WAV2LEV: Predicting Levenshtein Edit Operation Sequences For Fine-Grained Estimation of Automatic Speech Recognition Error
语音识别 | 7.5/10
论文解读
Whisper-FEST: Single-Channel Far-Field Enhanced Speech-to-text without Parallel Data
语音识别 | 7.5/10
论文解读
Whisper-MLA: Reducing GPU Memory Consumption of ASR Models Based on MHA2MLA Conversion
语音识别 | 7.0/10
论文解读
Whisper: Courtside Edition - Enhancing ASR Performance through LLM-Driven Context Generation
语音识别 | 6.5/10
论文解读
WhisperPipe: A Resource-Efficient Streaming Architecture for Real-Time Automatic Speech Recognition
语音识别 | 6.5/10
论文解读
Windowed SummaryMixing: An Efficient Fine-Tuning of Self-Supervised Learning Models for Low-Resource Speech Recognition
语音识别 | 6.5/10
论文解读
Z-Scores: A Metric for Linguistically Assessing Disfluency Removal
模型评估 | 6.5/10
论文解读
RAS: a Reliability Oriented Metric for Automatic Speech Recognition
语音识别 | 7.5/10
论文解读
Advancing automatic speech recognition using feature fusion with self-supervised learning features: A case study on Fearless Steps Apollo corpus
语音识别 | 7.0/10
论文解读
DM-ASR: Diarization-aware Multi-speaker ASR with Large Language Models
说话人识别 | 8.0/10
论文解读
Identifying and typifying demographic unfairness in phoneme-level embeddings of self-supervised speech recognition models
语音识别 | 7.0/10
论文解读
"This Wasn't Made for Me": Recentering User Experience and Emotional Impact in the Evaluation of ASR Bias
语音识别 | 7.0/10
论文解读
Do LLM Decoders Listen Fairly? Benchmarking How Language Model Priors Shape Bias in Speech Recognition
语音识别 | 7.5/10
论文解读
Evaluation of Automatic Speech Recognition Using Generative Large Language Models
语音识别 | 7.5/10
论文解读
Aligning Stuttered-Speech Research with End-User Needs: Scoping Review, Survey, and Guidelines
1. **问题**：当前口吃语音技术研究与口吃者（PWS）及言语语言病理学家（SLP）的实际需求存在系统性脱节，研究重点、任务定义和评估方法未能充分以用户为中心。 2. **方法核心**：通过两部分结合分析：1）对228篇相关论文进行范围综述，提出研究任务分类法并分析研究现状；2）对70名利益相
论文解读
Enhancing ASR Performance in the Medical Domain for Dravidian Languages
这篇论文旨在解决达罗毗荼语言（Telugu和Kannada）在医疗领域自动语音识别（ASR）中面临的标注数据稀缺和语言形态复杂两大挑战。其核心方法是提出一个“置信度感知训练框架”，该框架通过一个混合置信度评分机制（结合静态的感知、声学相似性、WER分数和动态的模型熵），对混合了真实与合成语音的训练数
论文解读
Reducing the Offline-Streaming Gap for Unified ASR Transducer with Consistency Regularization
1. **问题**：训练一个既能高精度离线转录又能低延迟流式识别的统一ASR模型极具挑战性，传统方法在低延迟下性能会急剧下降。 2. **方法核心**：提出一个统一的Transducer框架，结合分块注意力（含右上下文）和动态块卷积（DCConv）来适配两种模式。核心创新是引入了模式一致性正则化损失
论文解读
Tadabur: A Large-Scale Quran Audio Dataset
1. **问题**：现有的古兰经语音数据集在规模、诵读者多样性、音频质量和标注深度上存在严重不足，限制了古兰经ASR、诵读者识别等任务的研究进展。 2. **方法核心**：提出Tadabur数据集及其构建流水线。流水线核心是“古兰经经文对齐模块”（AAM），它结合WhisperX进行初步转录，再利用