Improved spoken language translation using n-best speech recognition hypotheses
Improved spoken language translation using n-best speech recognition hypotheses
复制标题
使用 n 种最佳语音识别假设改进口语翻译
DOI：
10.21437/interspeech.2004-53
复制
发表时间：
2004
期刊：
影响因子：
--
通讯作者：
W. Lo
中科院分区：
文献类型：
--
作者：
Ruiqiang Zhang;G. Kikui;H. Yamamoto;F. Soong;Taro Watanabe;E. Sumita;W. Lo
我们旨在演示使用N -最佳语音识别假设来提高语音翻译性能的效果。采用对数线性模型对候选译文进行重新评分，该模型综合了语音识别和统计机器翻译的特征。通过优化客观可测量但主观相关的翻译质量度量来估计模型参数。实验结果表明，所提出的N-best方法比传统的单一最佳方法提高了翻译质量。这些改进得到了几个自动翻译评估指标的一致证实。
We intended to demonstrate the effect of using N -best speech recognition hypotheses for improving speech translation performance. A log-linear model, which integrated features from speech recognition and statistical machine translation, was used to rescore the translation candidates. Model parameters were estimated by optimizing an objectively measurable but subjectively relevant translation quality metric. Experimental results have shown that the proposed N -best approach improved translation quality over the conventional single-best approach. The improvements were confirmed consistently by several automatic translation evaluation metrics.