Performance investigation of joint source-channel coding of speech for Persian dialectic, using MELP vocoder and MAP decoding
Ahmad Kasaeyan, Mohammad Hossein Moghaddam, Ghasem Assarzadeh · 2016
Mixed-excitation linear prediction (MELP) is a United States department of defense speech coding standard used mainly in military applications and satellite communications, secure voice, and secure radio devices. MELP vocoder could provide an acceptable voice quality over prone erroneous channels in various environments by exploiting redundancy statistics of voice data. This paper explains performance investigation of the MELP vocoding algorithm in conjunction with convolutional codes, and maximum a posteriori techniques in both hard and soft decoding regimes which utilize speech redundancy statistics of Persian language. We have evaluated our results according to weighted spectral distortion (WSD) and ITU criterion, perceptual evaluation of speech quality (PESQ).