Fully Convolutional ASR for Less-Resourced Endangered Languages
Fully Convolutional ASR for Less-Resourced Endangered Languages
复制标题
DOI：
--
复制
发表时间：
2020-05
期刊：
影响因子：
--
通讯作者：
Bao Thai;Robert Jimerson;R. Ptucha;Emily Tucker Prud'hommeaux
中科院分区：
文献类型：
--
作者：
Bao Thai;Robert Jimerson;R. Ptucha;Emily Tucker Prud'hommeaux
深度学习在自动语音识别 (ASR) 中的应用使训练数据丰富的语言的准确性得到了显着提高，但训练资源有限的语言尚未看到如此大规模的准确性提高。在本文中，我们将 ASR 声学建模的全卷积方法与各种已建立的声学建模方法进行了比较。我们在塞内卡（北美使用的一种资源匮乏的濒危语言）上评估了我们的方法。我们的方法产生的单词错误率比使用标准 GMM-HMM 方法和已建立的深度神经方法报告的错误率低 40%，并且训练时间大幅减少。这些结果显示了像 Seneca 这样既濒临灭绝又缺乏大量文档的语言的特殊前景。
The application of deep learning to automatic speech recognition (ASR) has yielded dramatic accuracy increases for languages with abundant training data, but languages with limited training resources have yet to see accuracy improvements on this scale. In this paper, we compare a fully convolutional approach for acoustic modelling in ASR with a variety of established acoustic modeling approaches. We evaluate our method on Seneca, a low-resource endangered language spoken in North America. Our method yields word error rates up to 40% lower than those reported using both standard GMM-HMM approaches and established deep neural methods, with a substantial reduction in training time. These results show particular promise for languages like Seneca that are both endangered and lack extensive documentation.