Techniques for effective vocabulary selection
计算与语言
2007-05-23 v1 人工智能
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.
引用
@article{arxiv.cs/0306022,
title = {Techniques for effective vocabulary selection},
author = {Anand Venkataraman and Wen Wang},
journal= {arXiv preprint arXiv:cs/0306022},
year = {2007}
}
备注
4 pages. To appear Proc. Eurospeech 2003, Geneva