arXivDaily arXiv每日学术速递 周一至周五更新
arXiv cs/0306022cs.CLcs.AI
收藏

有效词汇选择技巧

Techniques for effective vocabulary selection

Anand Venkataraman, Wen Wang

摘要:

连续语音识别(CSR)系统的词汇量是决定其性能的重要因素。在本文中,我们提出了三个原则性的方法来选择一个特定领域的目标词汇量之间的权衡目标的词汇表(OOV)率和词汇量的大小。我们评估这些方法对一个特设的基线策略。结果以OOV率图的形式呈现,这些图针对每种技术的词汇量增加而绘制。

英文摘要:

The vocabulary of a continuous speech recognition (CSR) system is a significant factor in determining its performance. In this paper, we present three principled approaches to select the target vocabulary for a particular domain by trading off between the target out-of-vocabulary (OOV) rate and vocabulary size. We evaluate these approaches against an ad-hoc baseline strategy. Results are presented in the form of OOV rate graphs plotted against increasing vocabulary size for each technique.

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。

↑