资源列表
Research_on_DTW-based_speech_recognition_for_voice
- 研究了将语音识别中的 ( 动态时间规整) 算法用于声纹鉴别的技术。通过引入 DTW Dynamic Time Warping, , 由所给的有限个样本建立最大相似于样本点的样本域, 计算被测样本的相似度。该算法提高了语 “样本域”的概念 音鉴别( 区分不同发音者) 的效率。有限人数的实验结果显示该算法辨伪率为 ( 人次) , 识别率 (98.75 400 81~93 80 人次) 。-Will be studied in speech recognition (dynamic
analoge_high_pass
- In oreder to compare Impulse Invariance method charactristics for transformation analogue filters to digital ones in highpass and bandpass and lowpass formats,here are 6 M-files that simulate a third order butterworth filter analyizing impulse respon
dist_IS
- Itakura-Saito (IS) Objective Speech Quality Measure
dist_LLR
- Log Likelihood Ratio (LLR) Objective Speech Quality Measure
wtd588d
- 这个是WT588D的语音处理程序. 一线控制发音.相当的不错-This is the voice WT588D process. First-line control of pronunciation. Very good
Geist
- An application ment to make use of voice recognition and emulation engines in java. So far only the emulation is implemented
VQ-kmmr
- Vecter quantification by kmmr
ftrack
- Tracks the formants through a method of averaging different resolutions of time frequency lpc based spectrograms
Pitchautv
- Pitch estimation using the autocorrelation method
ICA_MIX
- code to calculate the gaussian and generate the mixing of two signals in ICA
CENTERING
- code for centering process in ICA
whitening
- code for whitening in ICA