• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>IEEE International Conference on Intelligent Techniques in Control, Optimization and Signal Processing >Performance analysis and comparison of telephone speech enhancement algorithm for HOH listeners using OMAP processor based embedded systems
【24h】

Performance analysis and comparison of telephone speech enhancement algorithm for HOH listeners using OMAP processor based embedded systems

机译：基于OMAP处理器的嵌入式系统的HOH侦听器电话语音增强算法的性能分析与比较

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Hearing impairment detrimentally affects communication over the telephone line. Since phone lines reduce bandwidth and dynamic range, the poor quality speech signal can cause hard of hearing (HOH) listeners to experience extreme frustration and inefficient communication. The HOH listeners cannot hear the consonants having low energy, based on their degree of hearing loss. A telephone speech enhancement algorithm (TSEA) is developed to apply gain, to low amplitude signals by means of a threshold and leaving high amplitude signals. Applying gain to high amplitude signal will cause pain to the listeners. Performance analysis is done on the hardware implementation of telephone speech enhancement algorithm on Beagle Board XM and Panda Board. Comparison is done between the performances of speech enhancement algorithm across these two different OMAP Processor based embedded boards built using ARM processor. Initially an embedded system is built with Beagle Board XM as hardware and Android OS on Linux kernel as software. Similarly an embedded system is built with Panda Board as hardware and Android OS on Linux kernel as software. Simulink model for telephone speech enhancement is developed and tested with Beagle Board XM and Panda board. A test for TSEA algorithm is conducted using hard of hearing(HOH) listeners to prove the effectiveness of the algorithm when running on Beagle Board XM and Panda Board. These development boards accept analog audio signals, processes them using TSEA by executing speech enhancement algorithm and outputs the processed signal for transmission. The performance of the algorithm on both the boards may be analyzed for dynamic range in dB, hearing ability, SNR, power consumption and execution. It was implemented on both the boards and identified that the quality of the output speech signal is improved with this algorithm. It was also proved that the performance of this algorithm with Panda Board was better than with Beagle Board-XM.
机译：听证障碍不利地影响电话线路的沟通。由于电话线减少了带宽和动态范围，因此质量差的语音信号可能导致听力（HOH）听众造成极端挫折和效率低下的通信。由于其听力损失，HOH听众无法听到具有低能量的辅音。开发了电话语音增强算法（TSEA）以通过阈值施加增益，以阈值施加低幅度信号并留下高幅度信号。将增益施加到高幅度信号会对听众引起疼痛。绩效分析是关于猎犬板XM和熊猫板上的电话语音增强算法的硬件实现。使用ARM处理器构建的基于两个不同的OMAP处理器的语音增强算法的性能之间进行了比较。最初，嵌入式系统是在Linux内核上的Beagle Board XM作为硬件和Android OS构建的。类似地，嵌入式系统是用熊猫板作为硬件和Android OS在Linux内核上的硬件和Android OS构建。使用Beagle Board XM和Panda Loard开发和测试了用于电话语音增强的Simulink模型。使用听力（HOH）听众进行了对TSEA算法的测试，以证明在猎犬板XM和熊猫板上运行时算法的有效性。这些开发板接受模拟音频信号，通过执行语音增强算法使用TSEA来处理它们，并输出处理后的信号进行传输。可以分析DB，听力能力，SNR，功耗和执行中的动态范围对两个板上的算法的性能。它在两个板上实现并确定了该算法改进了输出语音信号的质量。还有人证明，这种算法与熊猫板的表现优于比猎犬板-XM更好。

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