一种改进的伪逆控制方案在DIVA模型中的应用

语音生成与获取是一个涉及大脑诸多部位的复杂认知过程,这个过程包括一种从依照句法和语义组织句子或短语的表述一直延伸到音素产生的分层结构。DIVA(directions into velocities of artculators)模型,是一种关于语音生成与获取后描述相关处理过程的数学模型,也是一种为了生成单词、音节或者音素,用来控制模拟声道运动的自适应网络模型。在当今真正具有生物学意义的语音生成和获取的神经网络模型中,DIVA模型的定义和测试相对而言是最彻底的,并且是唯一一种应用伪逆控制方案的模型。文中引入基于零空间的伪逆算法,对DIVA模型中的伪逆控制求解算法进行改进,从而更加精确地获得了DIVA模型的相应参数,提高了DIVA模型的鲁棒性。

An Improved Pseudo-inverse Control Scheme Applying to DIVA Model

Speech acquisition and production is a complex cognitive process of involving many parts of brain.This process includes organizing the sentences and phrases expression according to the syntactic and semantic.The process extends to the phonemes produces layered structure.DIVA(directions into velocities of artculators)is a mathematical model of the processes behind speech acquisition and production.It is also an adaptive neural network that learns to control movements of a simulated vocal tract in order to produce words,syllables or phonemes.In current real biologically meaningful speech acquisition and production neural network models,the definition and test of DIVA is relatively the most thorough and is the only model which apply pseudo-inverse control scheme.This article introduces the pseudo-inverse algorithm based on zero space,which improves pseudo-inverse control scheme of DIVA in order to obtain corresponding parameters of DIVA model more accurately.The proposed algorithm improves the robustness of DIVA model.