为满足在高误码率的窄带信道上进行语音通信的需求 ,研究了一种适用于甚低速率语音通信的抗误码参数估值算法。基于一定的解码状态 ,声码器通过最小均方误差(MMSE)估计的方法估计最优参数 ,充分降低信道误码对重建语音质量的影响。对于解码状态参数 ,通过计算最大后验转移概率的方法作最佳估计 ,并给出了一种简化的估计方法。这种抗误码算法计算量小 ,算法复杂度低。仿真结果表明 ,在不同误码率下用该算法恢复出的重建语音 ,不论客观评价或是主观评价 ,其质量都要优于传统的帧删除掩盖方法 In order to meet the requirement of voice communications over narrowband channels with high bit error rates, an anti-error parameter estimation algorithm suitable for very low rate speech communication is studied. Based on a certain decoding status, vocoder estimates the optimal parameters by means of minimum mean square error (MMSE) estimation, which reduces the impact of channel error on reconstructed voice quality. For the decoding state parameters, the best estimate is made by calculating the maximum posteriori transfer probability, and a simplified estimation method is given. This kind of anti-error code algorithm has a small amount of computation and low algorithm complexity. The simulation results show that the quality of the reconstructed speech recovered by this algorithm under different bit error rates is better than the traditional method of removing frames by either objective or subjective evaluation