搜索资源列表
Research_0n_Speech_Cepstral_Features
- 该文在研究基于线性预测倒谱和非线性MEL刻度倒谱特征的基础上,研究了LPCC和MFCC参数提取的算法原理及提取算法,提出了一级、二级差分倒谱特征参数的提取算法。识别实验验证了MFCC参数的鲁棒性优于LPCC参数。-In this paper, research is based on linear prediction and nonlinear MEL Cepstrum Cepstrum scale, based on studies of LPCC and MFCC parameter ex
Speech-recognition-in-MATLAB-
- 介绍了一种基于MATLAB的多个特定人连接词语音识别的方法,并提出了在进行端点检测时,引入平均的概念能进一步提高识别率。此设计是以LPCC系数、DTW算法为核心的基于图形界面的设计。-A Based on MATLAB more specific conjunctions speech recognition, and during endpoint detection, the introduction of the concept of average to further improve