从检测到预判：口吃的连续 onset 预测
From Detection To Anticipation: Continuous Onset Forecasting For Stuttered Speech
AI总结：
本研究将口吃预测建模为连续因果时间-事件预测，基于Whisper-large-v3和因果Conformer模型，在FluencyBank Timestamped数据集上实现0.63的AUROC，为预判口吃系统提供了基础。
摘要：
自动口吃系统主要在言语不流畅发生后检测到它们，这限制了其在预判干预中的应用。我们通过将口吃预测建模为连续因果时间-事件预测来解决这一差距，其中模型仅使用先前的语音每100毫秒估计即将发生的口吃 onset 的风险。为支持该任务，我们为36位说话者的FluencyBank Timestamped数据集补充了经人工审核的重复、延长和阻塞数据，得到2872个预测事件。随后，我们使用具有367万个可训练参数的紧凑因果Conformer模型对冻结的Whisper-large-v3语音表示的时间演化进行建模。所提出的方法达到了0.63的AUROC。时间分析进一步表明，显式建模先前的语音状态优于单纯增加静态声学窗口，而离散时间生存公式可使用单个模型在300毫秒至1.5秒范围内提供连贯的预测。这些结果提供了初步证据，表明口吃 onset 之前的语音中存在可测量的时间信息，并为未来的预判口吃系统建立了框架。
英文摘要：
Automatic stuttering systems primarily detect dysfluencies after they have occurred, limiting their use for anticipatory intervention. We address this gap by formulating stuttering prediction as continuous causal time-to-event forecasting, where a model estimates the risk of an upcoming stuttering onset every 100 ms using only preceding speech. To support this task, we augment FluencyBank Timestamped for 36 speakers with manually reviewed repetitions, prolongations, and blocks, yielding 2,872 forecasting events. We then model the temporal evolution of frozen Whisper-large-v3 speech representations using a compact causal Conformer with 3.67M trainable parameters. The proposed approach achieves an AUROC of 0.63. Temporal analysis further shows that explicitly modeling preceding speech states outperforms simply increasing the static acoustic window, while a discrete-time survival formulation provides coherent forecasts from 300 ms to 1.5 s using a single model. These results provide initial evidence that stuttering onset is preceded by measurable temporal information in speech and establish a framework for future anticipatory stuttering systems.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
