关于对话语流和流畅性塑造特征的令牌级建模难度
On the Difficulty of Token-Level Modeling of Dysfluency and Fluency Shaping Artifacts
AI总结：
本文提出一种轻量级适应方法，用于提升ASR对语流和流畅性修改的识别能力，同时揭示多语言端到端系统在处理德语数据时的局限性。
摘要：
自动转录含口吃语音仍是一个挑战，即使是现代端到端（E2E）自动语音识别（ASR）框架也是如此。语流和流畅性塑造特征常常被忽视，导致非逐字转录，具有有限的临床和研究价值。我们提出了一种参数高效的适应方法，将语流和流畅性修改作为特殊令牌纳入转录中，评估在模拟（LibriStutter，英语）和自然（KSoF，德语）口吃语音数据集上。为缓解ASR性能差异和对英语的偏见，我们引入了多步骤微调策略，结合语言自适应预训练。令牌化分析进一步突显了分词器的英语中心偏见，这给改进德语数据的性能带来了挑战。我们的发现证明了轻量级适应技术在语流意识ASR中的有效性，同时揭示了多语言端到端系统的关键限制。
英文摘要：
Automatic transcription of stuttered speech remains a challenge, even for modern end-to-end (E2E) automatic speech recognition (ASR) frameworks. Dysfluencies and fluency-shaping artifacts are often overlooked, resulting in non-verbatim transcriptions with limited clinical and research value. We propose a parameter-efficient adaptation method to decode dysfluencies and fluency modifications as special tokens within transcriptions, evaluated on simulated (LibriStutter, English) and natural (KSoF, German) stuttered speech datasets. To mitigate ASR performance disparities and bias towards English, we introduce a multi-step fine-tuning strategy with language-adaptive pretraining. Tokenization analysis further highlights the tokenizer's English-centric bias, which poses challenges for improving performance on German data. Our findings demonstrate the effectiveness of lightweight adaptation techniques for dysfluency-aware ASR while exposing key limitations in multilingual E2E 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。
