Application of triphone clustering in acoustic modeling for continuous speech recognition in Bengali
Application of triphone clustering in acoustic modeling for continuous speech recognition in Bengali
复制标题
三音素聚类在孟加拉语连续语音识别声学建模中的应用
DOI：
10.1109/icpr.2008.4761657
复制
发表时间：
2008
期刊：
影响因子：
--
通讯作者：
A. Basu
中科院分区：
文献类型：
--
作者：
Pratyush Banerjee;Gaurav Garg;Pabitra Mitra;A. Basu
声学模型的性能高度反映了任何连续语音识别系统的整体性能。因此，准确和鲁棒的声学模型的生成是令人满意的识别性能的关键。由于音素被发现根据在特定单词内出现的位置而变化，因此上下文信息在语音信号的声学建模中是最重要的。在本文中，我们期待在孟加拉语的连续语音识别的背景下，基于三音子的声学模型的效果单音子的声学模型。考虑到孟加拉语中缺乏用于基于三音素的声学建模的训练资源，我们在本文中还描述了使用基于决策树的技术生成三音素集群的方法。这些三音子集群，然后被用来生成绑定状态的三音子为基础的声学模型中使用的连续语音识别器。
The performance of the acoustic models is highly reflective on the overall performance of any continuous speech recognition system. Hence generation of an accurate and robust acoustic model holds the key to satisfactory recognition performance. As phones are found to vary according to the position of occurrence within a particular word, context information is of prime importance in acoustic modeling of phonetic signals. In this paper we look at the effect of triphone-based acoustic modeling over monophone based acoustic models in the context of continuous speech recognition in Bengali. Keeping in mind the lack of training resources for triphone-based acoustic modeling in Bengali, we have also described herein, the method of generating triphone clusters using decision tree based techniques. These triphone clusters have then been used to generate tied-state triphone based acoustic models to be used in a continuous speech recognizer.
通过血管紧张素抑制剂的“瞬时脉冲给药”抑制高血压的发作来研究高血压的消退
DOI：
--
发表时间：
2008
期刊：
影响因子：
--
作者：
石黒 喜美子;篠村 裕之;坂巻 裕介;林 香;伊藤 裕
通讯作者：
伊藤 裕