首页> 外文会议>Proceedings 2013 International Conference on Mechatronic Sciences, Electric Engineering and Computer >Realization of speech processing and recognition system based on digital signal processor
【24h】

Realization of speech processing and recognition system based on digital signal processor

机译:基于数字信号处理器的语音处理识别系统的实现

获取原文
获取原文并翻译 | 示例

摘要

16-bit fixed-point high-speed chip TMS320VC5402 and audio interface chip TLV320AIC10 are adopted in this paper, by which collecting, filtering and compressing data are processed. Final data are put into computer through serial port and received by MATLAB to complete playback, coding, recognition and simulation. Floating-point operation is realized on fixed-point digital signal processor, which has higher precision in calculation and wider signal processing dynamic range comparing with fixed-point realization. The experimental results show that rate of continuous word speech recognition is more than 92%. The new system has advantage of small size, low cost and high scalability, which can be applied to many specific occasions, such as the speech controlling toys and access control system.
机译:本文采用16位定点高速芯片TMS320VC5402和音频接口芯片TLV320AIC10,对数据进行采集,滤波和压缩处理。最终数据通过串行端口输入计算机并由MATLAB接收,以完成回放,编码,识别和仿真。浮点运算是在定点数字信号处理器上实现的,与定点实现相比,浮点运算具有更高的计算精度和更宽的信号处理动态范围。实验结果表明,连续单词语音识别率达到92%以上。该新系统具有体积小,成本低和可扩展性高的优点,可以应用于许多特定场合,例如语音控制玩具和门禁系统。

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