【24h】

A DC motor speed control using the LPC-ANFIS speech recognition system

机译:使用LPC-ANFIS语音识别系统的直流电动机速度控制

获取原文
获取原文并翻译 | 示例

摘要

The aim of this research is to design an implementation of the speech recognition system to control the speed of a DC motor. The Linear Predictive Coding (LPC) method is used in the speed recognition system, tuned by the Adaptive Neuro-Fuzzy Inference Systems (ANFIS) method. There are 5 (five) samples of voice signals in Bahasa Indonesia recognized by this system, i.e.: “Nyala”, “Lambat”, “Sedang”, “Cepat” and “Mati”. Every voice signal is repeated 5 (five) times until as many as 25 samples are recorded. Their voice characteristics are extracted using the LPC method represented by the LPC coefficients stored in a database system. The ANFIS method is implemented in 50 iterations to tune and to train the LPC coefficients until the least error, i.e. 0,00012446 is obtained. Voice samples originated from the internal database system are 83% successfully recognized by this system. However; samples extracted from the human voice signals of different persons - different sex from the person whose voice signals are recorded in the database system, and from various ages - are only 78,8% successfully recognized by the system. The output of the speech recognition system is coded into the ASCII Codes and converted into the PWM signal to control the speed of a DC motor.
机译:这项研究的目的是设计语音识别系统的实现,以控制直流电动机的速度。线性预测编码(LPC)方法用于速度识别系统中,由自适应神经模糊推理系统(ANFIS)方法进行调整。该系统在印度尼西亚语中有5(五个)语音信号样本,即:“尼亚拉”,“兰巴特”,“雪邦”,“塞帕特”和“马蒂”。每个语音信号重复5(五)次,直到记录了多达25个样本。使用由存储在数据库系统中的LPC系数表示的LPC方法提取其语音特性。以50次迭代实施ANFIS方法,以调整和训练LPC系数,直到获得最小误差(即0,00012446)为止。源自内部数据库系统的语音样本已被该系统成功识别83%。然而;从不同人的人的语音信号中提取的样本(该人的性别与在数据库系统中记录语音信号的人的性别不同,并且来自不同年龄)仅被系统成功识别了78.8%。语音识别系统的输出被编码为ASCII码,然后转换为PWM信号以控制直流电动机的速度。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

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