|
会话语音识别的串联连接特征提取
Tandem Connectionist Feature Extraction for Conversational Speech Recognition |
|
| |
| 课程网址: |
http://videolectures.net/mlmi04ch_zhu_tcfec/
|
| 主讲教师: |
Qifeng Zhu
|
| 开课单位: |
国际购物中心协会
|
| 开课时间: |
2007-02-25
|
| 课程语种: |
英语
|
| 中文简介: |
多层感知器（MLP）可以以多种方式用于自动语音识别。如Hermansky等人在许多出版物中所描述的，该工具在过去几年中的特定应用是Tandem方法。在这里，我们讨论了用于Tandem方法的基于MLP的特征的特征，并总结了它们在会话语音识别中的应用的报告。本文表明，MLP变换产生具有规则分布的变量，可以通过使用对数进一步修改，以使分布更容易通过高斯HMM进行建模。可以在不增加特征尺寸的情况下容易地组合这些特征的两个或更多个矢量。我们还报告了识别结果，这些结果表明，MLP功能可以显着提高NIST 2001 Hub 5评估集的识别性能，其中包括使用Switchboard Corpus培训的模型，即使对于包含MMIE培训和其他增强功能的复杂系统也是如此。
|
| 课程简介: |
Multi-Layer Perceptrons (MLPs) can be used in automatic speech recognition in many ways. A particular application of this tool over the last few years has been the Tandem approach, as described by Hermansky et al in a number of publications. Here we discuss the characteristics of the MLP-based features used for the Tandem approach, and conclude with a report on their application to conversational speech recognition. The paper shows that MLP transformations yield variables that have regular distributions, which can be further modified by using logarithm to make the distribution easier to model by a Gaussian-HMM. Two or more vectors of these features can easily be combined without increasing the feature dimension. We also report recognition results that show that MLP features can significantly improve recognition performance for the NIST 2001 Hub-5 evaluation set with models trained on the Switchboard Corpus, even for complex systems incorporating MMIE training and other enhancements.
|
| 关 键 词: |
多层感知器; 语音识别; 规则分布
|
| 课程来源: |
视频讲座网
|
| 最后编审: |
2019-07-02：cwx
|
| 阅读次数: |
87
|
|
图 书 馆