arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2101.06396eess.AScs.LGcs.SD
收藏

基于不确定性模型的非母语(L2)英语发音错误检测

Mispronunciation Detection in Non-native (L2) English with Uncertainty Modeling

Amazon Speech(亚马逊语音);Gdansk University of Technology(格但斯克工业大学)

Daniel Korzekwa, Jaime Lorenzo-Trueba, Szymon Zaporowski, Shira Calamaro, Thomas Drugman, Bozena Kostek

摘要:

在语言学习中自动检测发音错误的一种常见方法是识别学生产生的音素,并将其与母语者的预期发音进行比较。这种方法有两个简单的假设:a)音素可以从语音中高精度地识别出来,b)一个句子的发音只有一种正确的方式。这些假设并不总是成立,这可能导致大量的错误发音警报。我们提出了一种新的方法来克服这个问题的基础上两个原则:a)考虑到不确定性的自动音素识别步骤,b)占的事实,可能有多个有效的发音。我们评估的非母语(L2)的德语,意大利语和波兰语的英语语音模型,在那里它被证明可以增加高达18%(相对)的常见方法相比,检测发音错误的精度。

英文摘要:

A common approach to the automatic detection of mispronunciation in language learning is to recognize the phonemes produced by a student and compare it to the expected pronunciation of a native speaker. This approach makes two simplifying assumptions: a) phonemes can be recognized from speech with high accuracy, b) there is a single correct way for a sentence to be pronounced. These assumptions do not always hold, which can result in a significant amount of false mispronunciation alarms. We propose a novel approach to overcome this problem based on two principles: a) taking into account uncertainty in the automatic phoneme recognition step, b) accounting for the fact that there may be multiple valid pronunciations. We evaluate the model on non-native (L2) English speech of German, Italian and Polish speakers, where it is shown to increase the precision of detecting mispronunciations by up to 18% (relative) compared to the common approach.

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。

↑