首页> 外国专利> Method for recognition of N different successions of words in a speech signal

Method for recognition of N different successions of words in a speech signal

机译:语音信号中的n个不同的单词连续识别方法

摘要

In order to be able to determine, in a known method for recognition of a succession of words in a speech signal, not only the succession of words with the greatest similarity, but also a specific number of further successions of words with decreasing similarity, a specific structure is indicated for the storage spaces in the memory with the backtracking addresses, as is a measure for generating the information for these storage spaces. The computing effort additionally required for determining these further successions of words is low in comparison with the normal computing effort of the recognition method. IMAGE
机译:为了能够在用于识别语音信号中的单词连续的已知方法中,不仅确定具有最大相似度的单词的连续,而且还确定特定数量的具有相似度递减的单词的进一步连续,指示具有回溯地址的存储器中存储空间的特定结构,作为用于生成这些存储空间的信息的措施。与识别方法的正常计算工作相比,确定单词的这些进一步的连续所需要的额外计算工作量少。 <图像>

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号