评估判别式与生成式端到端语音增强模型对音节重音保留的影响
Evaluating the Impact of Discriminative and Generative E2E Speech Enhancement Models on Syllable Stress Preservation
AI总结：
该研究针对噪声环境下重音检测鲁棒性不足的问题，考察判别式与生成式端到端语音增强模型对音节重音保留的影响，结合机器检测与人类感知实验验证生成式模型结合启发式特征的鲁棒性。
摘要：
自动音节重音检测是面向语言学习者的计算机辅助语言学习（CALL）系统的核心组件。当前的重音检测模型通常在纯净语音上训练，在背景噪声普遍存在的真实场景中可能缺乏鲁棒性。为解决这一问题，可采用旨在通过去噪增强语音的语音增强（SE）模型，但这类模型对音节重音模式保留的影响尚未得到充分研究。\n本研究考察了分别代表判别式与生成式建模路径的不同SE模型，在噪声环境下对音节重音检测的影响。我们将这些模型应用于信噪比（SNR）在0至20 dB范围内的语音数据，评估其维持重音模式的效果。此外，我们探索了不同的特征集，以确定哪些特征最适合在噪声中捕捉重音模式。为进一步理解SE模型的影响，我们开展了一项人类感知研究，对比SE增强语音与纯净语音中感知到的重音模式，揭示这些模型在听众感知层面保留音节重音的效果。\n实验在来自德语和意大利语非母语使用者的英语语音数据上进行。结果表明，当使用启发式特征时，生成式SE模型下的重音检测性能具备鲁棒性。此外，感知研究的观察结果与所有SE模型下的重音检测结果一致。
英文摘要：
Automatic syllable stress detection is a crucial component in Computer-Assisted Language Learning (CALL) systems for language learners. Current stress detection models are typically trained on clean speech, which may not be robust in real-world scenarios where background noise is prevalent. To address this, speech enhancement (SE) models, designed to enhance speech by removing noise, might be employed, but their impact on preserving syllable stress patterns is not well studied. This study examines how different SE models, representing discriminative and generative modeling approaches, affect syllable stress detection under noisy conditions. We assess these models by applying them to speech data with varying signal-to-noise ratios (SNRs) from 0 to 20 dB, and evaluating their effectiveness in maintaining stress patterns. Additionally, we explore different feature sets to determine which ones are most effective for capturing stress patterns amidst noise. To further understand the impact of SE models, a human-based perceptual study is conducted to compare the perceived stress patterns in SE-enhanced speech with those in clean speech, providing insights into how well these models preserve syllable stress as perceived by listeners. Experiments are performed on English speech data from non-native speakers of German and Italian. And the results reveal that the stress detection performance is robust with the generative SE models when heuristic features are used. Also, the observations from the perceptual study are consistent with the stress detection outcomes under all SE models.
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。
