基于TMS320F2812的语音检测系统设计

Design of Speech Detection System Based on TMS320F2812

语音静噪是以语音检测为基础实现无语音时屏蔽系统输出的一项功能。由于短波电台接收机存在诸多噪声干扰,致使语音静噪较难实现。本文设计实现一种以TMS320F2812DSP芯片为核心的语音检测系统,对短波电台接收到的信号进行分析和处理,并应用短时自相关算法进行语音检测。如果检测存在语音信号则通过DA把语音转换输出。通过在实际应用环境中测试表明该语音检测系统结构简单,语音检测准确度高,可以有效解决短波电台的语音静噪问题。

Speech noise silencing is testing voice-based to achieve shield the system output function when there is no speech. As the short-wave radio receiver there is a lot of noise, with the result that more difficult to achieve noise-cancellation, in this paper,designed to achieve a voice detection system which make the TMS320F2812 DSP chip as its core. This system can analysis and process the signal received by the shortwave radio and detect the voice by the short-time auto-correlation. If speech signal is detected, then it can be transformed to output through DA. Through the practical application of the test environment show that the voice detection system is simple, highly accurate detection of voice can be an effective solution to short-wave radio voice of the noise silencing.