arXiv cs.CL· Lucky Onyekwelu-Udoka, Md Shafiqul Islam·· 12 小时前AI 评分29
研究：ASR 式转写规范化会移除对话行为分类所需的有用语用信号
ASR-Style Transcript Normalization Removes Useful Pragmatic Signal for Dialogue Act Classification
AI 导读
研究评估九种转写条件、六种分类器（含 RoBERTa、BERT、GPT-2）与三个对话语料库，发现重度规范化在全部 15 个神经模型-语料组合上均降低 Macro-F1。
正文 · 原文
Abstract:Automatic speech recognition (ASR) pipelines often normalize transcripts by removing filled pauses and punctuation and expanding colloquial forms, potentially discarding cues useful for dialogue act classification. We evaluate nine transcript conditions across six classifiers (TF-IDF logistic regression, RoBERTa, BERT, ALBERT, GPT-2, and BiLSTM) and three conversational corpora (Switchboard, ICSI MRDA, and SLUE-HVB), supplemented by multi-seed experiments, normalization controls, and real-ASR evaluations. Across five random seeds, heavy normalization reduces Macro-F1 in all 15 neural model-corpus combinations under the original preprocessing protocols. Removal using the original filler and backchannel lexicon consistently degrades performance on MRDA and SLUE-HVB but produces mixed effects on Switchboard. Punctuation-aware TF-IDF experiments show that punctuation removal reduces Macro-F1 by 0.036 on Switchboard and 0.078 on MRDA, demonstrating that the original linear baseline's apparent insensitivity resulted from tokenization. Restricting filler removal to um, uh, and erm preserves degradation on MRDA but largely eliminates the SLUE-HVB effect, where the original lexicon removes class-defining backchannel expressions. These findings demonstrate that normalization effects depend on preprocessing definitions, feature extraction, and conversational context, supporting task-dependent rather than indiscriminate transcript normalization.
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.22934 [cs.CL] |
| (or arXiv:2610.22934v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.22934 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Md Shafiqul Islam [view email]
[v1]
Fri, 16 Oct 2026 04:51:37 UTC (17 KB)
来源：arXiv cs.CL · arxiv.org