资源列表
get_fundamental
- get the fundamental component of a voice signal.
houzhi
- 改程序主要功能：对采集到的带噪语音进行后置滤波处理-Change program main functions: to the rear to the speech with noise filtering processing
lpc2lpcc
- 改程序功能：主要完成对采集到的语音信号进行lpc2lpcc参数进行提取-The app features: mainly to complete the speech signals were collected lpc2lpcc parameters extraction
mfcc
- 改程序的功能：在进行语音识别时，完成对采集到的信号进行MFCC参数提取-The app features: when making the speech recognition, to complete the collected signal MFCC parameters extraction
vad
- 改程序的功能：在进行语音识别时，完成对采集到的语音信号进行端点检测-The app features: when making the speech recognition, complete the acquisition to the endpoint detection of speech signal
recognition-based-dtw
- 基于动态时间规整算法实现的孤立词识别程序-Dynamic time warping algorithm to achieve an isolated word recognition program
VAD
- 附件内容为从Webrtc程序中抠出来的VAD检测算法-Webrtc VAD code
sincosFixed
- 在语音中需要对其进行波束分析，提供的是关于波束图的画法-beam plot db
vqlbg
- 用LBG算法进行矢量量化器的设计,用分裂法获得初始码书-Using vector quantization LBG algorithm design, with split method to obtain the initial codebook
Speech-Recognition
- 语音识别室 是进来AI的热们方向，本例是一个语音识别的案例，希望可以在语音智能处理上有所帮助。-Speech Recognition Room AI is coming in the popular direction, in this case is a case of voice recognition, and hope to be able to help in the processing of voice intelligence.
blog_code
- shouxieshuzishibei神经网络-Numerical Identification of Neural Networks
spro-5.0
- inria给出的一个工具，主要用于提取语音识别的特征-Speech recognition features