资源列表
melp
- melp 2.4K 浮点算法 语音编解码程序-melp 2.4K floating-point arithmetic voice encoding and decoding procedures
noise_cancelation
- 对语音信号中叠加上不同的噪声，用NLMS方法对带噪语音进行消噪处理。-Different noise superimposed on the speech signal, the NLMS method for Noisy Speech Denoising.
MFCCPHMM
- 本实例程序是基于matlab的语音识别程序，通过提取传统的MFCC特征集，采用HMM分类器，作为对比实验是一个不错的选择。-This example the program matlab speech recognition program to extract MFCC feature set, the HMM classifier, as a comparative experiment is a good choice.
pcm2wav
- PCM编码(A律或U律)格式文件转换成WAV格式文件的VC6.0项目。-A VC6.0 project to covert PCM format(a-law/u-law) to WAV format.
add_noisem
- 把指定的噪声叠加到信号上去.有标准噪声库NOISEX-92，其中带有白噪声、办公室噪声、工厂噪声、汽车噪声、坦克噪声等等，在信号处理中往往需要把库中的噪声叠加到信号中去，而噪声的采样频率与纯信号的采样频率往往不一致，需要采样频率的校准。 -The specified noise superimposed to the signal up. Standard noise library NOISEX-92, with white noise, office noise, factory noi
GMM_SID
- 用于说话人识别（声纹识别）中训练过程和识别过程的高斯混合模型程序-GMM model for the training process or test process of speaker identification
score_tradeoff
- 计算说话人识别系统的全局EER，DCF和平均EER，DCF-Calculation of the overall situation of the speaker recognition system the EER, DCF and average EER, DCF
HMM
- 语音识别与合成的基本程序，分为特征提取、模型建立和语音识别，且包括特征补偿-Basic procedures for speech recognition and synthesis, divided into feature extraction, modeling and speech recognition, including feature compensation
DTMF2
- DTMF检测程序。定点计算，可移植到DSP上。实现按键音软解。-DTMF detection program. Sentinel, can be transferred to the DSP. Key tone soft solution.
TIMIT
- TIMIT的一部分语料库，不是很全，刚从网上下载的希望对大家有所帮助-Part of the corpus of TIMIT, not very full, just downloaded we want to be helpful.
nlms-Double-end-detection
- 基于NLMS算法的回声消除，具有双端检测功能-Echo cancellation nlms Double-end detection
nvmex_tool
- use cuda with matlab. this is for older versions of matlab and users that dont have the pct