Tools for the analysis of benchmark speech recognition tests

Tools for the analysis of benchmark speech recognition tests
复制标题

DOI：
10.1109/icassp.1990.115546
复制
发表时间：
1990-04
期刊：
International Conference on Acoustics, Speech, and Signal Processing
影响因子：
--
通讯作者：
D. S. Pallett;W. Fisher;Jonathan G. Fiscus
D. S. Pallett;W. Fisher;Jonathan G. Fiscus
中科院分区：
其他
文献类型：
--
作者：
D. S. Pallett;W. Fisher;Jonathan G. Fiscus

文献摘要

被引文献

相似文献

开发工具的基准语音识别系统测试的分析报告。一个发展是一个工具，实现两个统计显著性检验。另一项研究涉及DARPA/NIST评分软件（目前最小化基本单词错误类型的加权和）中目前使用的比对过程的替代方案。这个替代过程最大限度地减少了语音上的不可信性。开发这些工具的标准实现的目的是使系统开发人员可以统一使用这些工具。
The development of tools for the analysis of benchmark speech recognition system tests is reported. One development is a tool implementing two statistical significance tests. Another involves studies of an alternative to the alignment process presently used in the DARPA/NIST scoring software (which presently minimizes a weighted sum of elementary word error types). The alternative process minimizes a measure of phonological implausibility. The purpose in developing a standard implementation of these tools is to make these tools uniformly available to system developers.>