资源列表
wav
- 语音分割程序，可将一段语音自动分割成极端语音，包括端点检测等子程序。-Speech segmentation procedure, a voice can be automatically split into extreme voice, including endpoint detection routines.
source-code
- this fingerprint combination for privacy protection code only contain feature extraction-this is fingerprint combination for privacy protection code only contain feature extraction
PlaySound
- PlaySound播放WAV文件 PlaySound播放WAV文件-play wav using PlaySound play wav using PlaySound play wav using PlaySound
wav
- 利用向量内积 比对两个相似音频 找出相似点 算出相似度-The use of vector product of two similar audio than to identify similarities calculated similarity
jiyintiqu
- 程序主要是求一段音频数据的基音频率 jiyinzhouqi.m --------求基音周期 find_maxn.m --------求自相关函数的最大值 test2.wav --------程序中用到的示例wav数据文件-The program is mainly the fundamental frequency for an audio data Jiyinzhouqi.m- for pitch period Find_maxn.m- the maximum
234234234
- 声纹识别包括完整的声音预处理和特征计算过程-Voiceprint identifying comprises preprocessing and feature a complete calculation of the sound
g711DecodeTest
- g711解码函数,里面包含了对g711编码的语音进行解码的操作,适合初学者学习.-It is g711 audio decode function,It will help us to learn how to decode g711 audio.
HMM
- 结合实例把隐马尔科夫模型讲解的简单易懂，值得一看-Easy to understand with an example to explain the hidden Markov model, worth a visit
Sound-Effect-board
- 这份报告概述了项目修改语音信号与各种不同的效果。实现的效果是一个呼应,相移和残余阈值剪裁。的分析过程和技术参与实施这些技术也提供了一个洞察整个项目选择背后的原因。这份报告将细分一个语音信号的分解步骤,将段到频域,估计谱包络,操纵信号以不同的方式,然后回到回放的时间域。附录中提供一份项目以及示例输出结果。还提到可能的扩展项目,可能会增强其功能和团队的建议进行这些增强功能。-This report outlines a project to modify a speech signal with v
speaker_recognition
- 说话人识别是自动识别的过程,他在个人的基础信息包含在演讲。这种技术可以使用扬声器的声音来验证他们的身份和访问控制服务,如语音拨号,银行通过电话、电话购物、数据库访问服务、信息服务、语音邮件、安全控制领域的机密信息,和远程访问电脑。-Speaker recognition is the process of automatically recognizing who is speaking on the basis of individual information included in spe
basicsetup
- voice application, push to talk, overwiew for personal area network
recog
- The sign language recognition to create a to store the training image, the video capture infront the user image and each sign language assign the letter the preprocessing the image then image is convert into rgg to cbcr. The image is segmented fe