资源列表
ofa
- wma格式的音乐文件识别，基于musicdns的音频指纹识别技术，能为音乐文件产生唯一的puid
voice
- 最近几年语音算法的论文，主要以语音识别为主，对研究语音信息处理有很大帮助-Voice algorithm in recent years, mainly composed of speech recognition, speech information processing are of great help to research
MATLAB2
- 基于HMM的语音识别的资料，希望对大家有所帮助。
matlabyuyinshibiesuanfa
- speech recognition awhich include training and testing
speech_denoising
- 高斯噪声背景下的语音去噪，附加matlabUI界面(Speech denoising under Gauss noise background)
freetts-1.0.tar
- FreeTTS（第一部分）-FreeTTS Part one
SpeakerVoiceIdentifier-master
- 用C++完成的基于MFCC和GMM的说话人识别软件。(Speaker recognition software based on MFCC and GMM is completed with C++.)
train-samples
- training samples for training in speech recognition
GmmTrainTestLBG
- it is a Gaussian Mixture Model(MATLAB version) for training and testing with some list or file. and it has references in enclosed rar file
BinauralDNN
- This Matlab program implements the cochannel speech separation algorithm Yi Jiang, D. L. Wang, et al. (2014), "Binaural Classification for Reverberant Speech Segregation Using Deep Neural Networks," .
VirtualSound
- 该程序是对于虚拟声音的模拟，其中包括了动态声音和静态声音的模拟，并附有MIT的声音采集库，程序有GUI界面，各种音频都完全，直接运行。-The program is sound for virtual simulation, including dynamic and static sound of analog sound, along with the voice of MIT library collection, the program has GUI interface, a vari
CS-yuyin
- 自己搜集的关于语音方面的论文，CS，压缩感知，是最近的热门，打破了采样定律，结合CS的语音论文，这里给出，很全面。-His collection of papers about speech, CS, compressed sensing, is recently popular, breaking the law of sampling, combined with the voice of the CS papers, shown here, is very comprehensive.