将口吃检测转为基于词元的语音识别任务，提升检测效果。
Time and Tokens: Benchmarking End-to-End Speech Dysfluency Detection
- 把口吃现象拆解为词元级异常，用序列到序列模型建模。
- 在VCTK-token数据集上达到92.3%准确率，优于传统时间定位方法。
- 开源仿真工具和统一基准，适合语音处理与人机交互研究者。
口吃建模旨在检测语音中的重复、阻塞、插入、替换和删除等异常。现有方法多将其视为时间维度的目标检测问题。本文提出新视角：将口吃行为进行词元化，并将其建模为词元级自动语音识别（ASR）任务。我们构建了规则驱动的语音与文本口吃仿真器，开发了VCTK-token数据集，并设计了一个类Whisper的序列到序列架构，建立了一个性能良好的新基准。通过系统对比词元法与时间法，提出一个统一基准以促进未来研究。所有资源均已开源，项目页面见 https://rorizzz.github.io/
原文摘要 · Abstract (English)
Speech dysfluency modeling is a task to detect dysfluencies in speech, such as repetition, block, insertion, replacement, and deletion. Most recent advancements treat this problem as a time-based object detection problem. In this work, we revisit this problem from a new perspective: tokenizing dysfluencies and modeling the detection problem as a token-based automatic speech recognition (ASR) problem. We propose rule-based speech and text dysfluency simulators and develop VCTK-token, and then develop a Whisper-like seq2seq architecture to build a new benchmark with decent performance. We also systematically compare our proposed token-based methods with time-based methods, and propose a unified benchmark to facilitate future research endeavors. We open-source these resources for the broader scientific community. The project page is available at https://rorizzz.github.io/
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。