because supporting vectors are near to the border, before training speaker identification system based on svm using speech training assemble, it is needed to reduce the training assemble.
由于支持向量具有边界性,在利用语音训练集对基于支持向量机(svm)的说话人识别系统进行训练之前,需要对该训练集进行约简。
to solve the effect of channel changes on the performance of speaker identification system , apply the method of maximum a posteriori to specific channel compensation.
为了解决通道变化对说话人识别系统性能的影响,将最大后验概率方法应用到具体的通道补偿中。
an inter-speaker information ann method is proposed, analyzing gmm model speaker identification system performance, and a hybrid gmm/ann speaker identification system is structured.
通过分析gmm(高斯混合模型)的说话人辨认系统的性能,提出了一种捕捉不同说话人交互信息的人工神经网络(ann)方法,构成一个gmm/ann混合说话人辨认系统。
the experiments results of the closed-set text-independent speaker identification system indicate that the proposed models and algorithms improve identification accuracy.
闭集文本自由说话人辨认试验证明了提出的模型及其算法的正确性。