搜索资源列表
VAD-DTW-HMM
- 端点检测和基于DTW和HMM的孤立词识别和连续语音识别-endpoint detection and DTW-based HMM and isolated word recognition and Continuous Speech Recognition
Voice_activity_detection
- 采用能量门限的语音活动端点检测程序，计算复杂度低．-using energy threshold voice activities endpoint detection procedures, low computational complexity.
cdhmm
- matlab下实现的用于语音识别的一些算法，包括hmm参数提取，viterbi识别算法，mfcc参数提取算法，端点检测算法等，对于做语音识别的来说都是很有用的-under Matlab for the voice recognition algorithm, including hmm parameter extraction, Recognition Viterbi algorithm, mfcc parameter extraction algorithm, endpoint detecti
voice464
- 基于dtw算法的语音识别 1 ENFRAME.M和MELBANKM.M取自voicebox工具箱 2 vad.m实现端点检测 3 mfcc.m计算mfcc参数 4 dtw.m实现DTW算法[训练] dtw2.m实现DTW高效算法 5 testdtw.m测试程序-based speech recognition algorithm and a ENFRAME.M MELBANKM.M from v oicebox Toolbox 2 vad.m ach
duandianjiance
- 语音信号端点检测仿真，内容涉及短时能量及过零率的计算，自相关函数的的计算。-voice endpoint detection signal simulation, which relates to short-term energy and zero rate calculation, the autocorrelation function of the calculation.
hmmPlatform
- 语音分析平台（不包括HMM模型） 平台源码中提供了语音文件的提取和处理方法，时域分析（预处理/能量/振幅/过零/端点检测），频域分析（fft/功率谱/倒谱）和矢量量化的实现(搜集转发)-voice analysis platform (not including HMM) platform to provide a source document voice extraction and processing methods , time-domain analysis (pretreatm
dsp61a
- 语音识别在遥控系统中的应用设计介绍一种适合家电遥控器应用的语音识别算法，该算法使用双模块和两级端点检测方法，能有效地提高识别和稳健性；介绍利用该技术实现的一种新型学习型遥控器，展现了语音识别技术在家电领域的广阔前景-voice recognition in the remote control system of Design for a remote appliance application of speech recognition algorithm, The algorithm use
record
- 语音mfcc,用来录取人声，进行端点检测，分析人声的音量(Voice MFCC, used for voice retrieval, endpoint detection, analysis of voice volume)
matlab
- 用于基本的语音分析，包含端点检测，谱减法。(Used for basic speech analysis)
train
- 基于BP网络和多特征的语音端点检测，提取自相关函数最大值和频带方差作为BP网络输入，输出判断是否为语音信号。(A Speech Endpoint Detection Algorithm Based on BP Neural Network and Multiple Features)
用短时平均能量和短时平均过零率双门限检测VAD
- 用短时平均能量和过零率的特征参数对语音信号进行端点检测， 以消除静默帧和噪声帧的干扰
An
- 语音端点检测，一种基于MFCC的余弦值单门限值检测方法，适用于低SNR环境(Speech endpoint detection, a single threshold value detection method based on the cosine value of MFCC, suitable for low SNR environment)
哼歌识曲
- 通过语音分帧、端点检测、pitch提取、DTW算法，实现歌曲识别(Through voice framing, endpoint detection, pit extraction, DTW algorithm, song recognition is realized)
short_zero_crossing_rate
- 语音信号的短时过零率求取，此过程为语音信号的一个重要特征提取过程，可用于端点检测等过程。(This process is an important feature extraction process for speech signals and can be used for endpoint detection and other processes.)
基于DTW算法的语音识别系统
- 通过语音分帧、端点检测、特征提取、DTW算法，实现多字语音识别(Through speech segmentation, endpoint detection, feature extraction and DTW algorithm, multi word speech recognition is realized.)
matlab
- 这是一个具有语音的采集、读取、内插恢复、重采样，语音的时域参数的计算、端点的检测、基音周期的提取，语音的加噪、滤波及每次处理后语音的播放等功能的语音信号处理系统。(speech signal processing)
voice_wav_frequency
- matlab语音信号分析时域频域，加窗，降噪滤波，端点检测，时域特征和频域特征提取(Matlab speech signal analysis in time domain, frequency domain, windowing, denoising, filtering, endpoint detection, time domain feature and frequency domain feature extraction.)
lab
- 端点检测、基因频率检测、说话人性别识别、GUI界面(Speaker sex identification)
mfc
- matlab中简单语音处理，包括小波滤波，端点检测和mfcc参数提取，输出amf为所读取的语音文件24*1维的mfcc参数(Matlab simple voice processing, including wavelet filtering, endpoint detection and MFCC parameter extraction, output amf for the read voice file 24*1 dimensional MFCC parameters)
vad-master
- 调用webrtc中的VAD算法，可以实现录音文件的语音端点检测，时延7.8ms(Calling VAD algorithm in webrtc, voice endpoint detection of recording file can be realized with a delay of 7.8ms)