文档下载 文档下载导航
设为首页 | 加入收藏
搜索 请输入内容:  
 导航当前位置: 文档下载 > cepstral

cepstral

  • Eurospeech 2001- Scandinavia Split-band Perceptual Harmonic cepstral Coefficients as Acoust

    Split-band Perceptual Harmonic Cepstral Coefficientsas Acoustic Features for Speech Recognition Liang Gu and Kenneth Rose Department of Ele...

  • Mel Frequency cepstral Coefficients An Evaluation of Robustness of MP3 Encoded Music

    MelFrequencyCepstralCoef cients:AnEvaluationofRobustnessofMP3EncodedMusicSigurdurSigurdsson,KaareBrandtPetersenandTueLeh...

  • Eurospeech 2001- Scandinavia Split-band Perceptual Harmonic cepstral Coefficients as Acoust

    Split-band Perceptual Harmonic Cepstral Coefficientsas Acoustic Features for Speech Recognition Liang Gu and Kenneth Rose Department of Ele...

  • 噪声背景下连续语音信号分割的一种新方法

    ...eral different results derived from different endpoint detection methods based on fractal dimension,cepstral feature and HMM model,using a candidate selection approach to get the final boundary in order to se...

  • AUTOMATIC SEGMENTATION AND LABELING OF SPEECH RECORDED IN UNKNOWN NOISY CHANNEL ENVIRONMENT

    ...LSP Constrained Iterative Wiener ltering. In addition, the coupling of front-end processing such as cepstral mean subtraction in tandem with speech enhancement methods is considered for noisy channel compensa...

  • 一种基于SOPC的语音_文本转换技术

    ...展,为对语音信号进行处理及文字识别技术的实现提供了良好的技术基础和实现手段。语音 文本转换系统的核心模块是对复杂语音信号的处理与识别。系统采用LPC的方法对语音进行分析[1],从LPC系数获得加权的Cepstral系数以及Cepstral的时间微分[2],从而形成每一帧的特征矢量,经过对其的矢量量化可以得到观察序列。通过采用隐马尔可夫模型(HiddenMarkovModel,HMM)[3]进行基于音素的训练,训...

  • Acoustical and Environmental Robustness in Automatic Speech Recognition

    ...eady provided by existingsuccessful acoustical pre-processing algorithms such as codeword-dependent cepstral normalization (CDCN) [1]. Experimental tech-niques are proposed in the course of investigating the ...

  • 基于GMM_UBM的稳健说话人识别系统

    ... Active Detection, VAD)非常重要。本文提出了基于子带能量和基频提取的VAD算法,较好的过滤了录音中的静音和噪声。 我们在系统前端的频谱特征提取阶段使用了倒谱均值方差归一化方法(Cepstral Mean and Variance Normalization, CMVN)去除通道不匹配和加性噪声干扰。虽然近年来有很多特征归一化方法,比如非参数化方法的直方图规整(Histogram[6]和参数...

  • sphinx4-作用

    ...(预加重), Hamming window, FFT, Mel frequency filter bank(mel频率过滤), discrete cosine transform(离散cos变换), cepstral(同态谱/倒频谱) mean normalization (标准化), and feature extraction (特征提取) of cepstra, delta cepstra, double ...

  • HTK孤立词识别回顾

    ...tion for windowing frames # Pre-emphasis coefficient # Number of filterbank channels # Length of cepstral liftering # The Endtargetlist.txt说明抽取源文件路径和目标文件的保存路径,内容如下:data/train/sig/yes01.sig data/train/m...

  • Multimodal Person Verification from Video Sequences

    ...ysis and independent component analysis techniques are used for face verification and mel-frequency-cepstral coefficients are used for speaker verification. The matching scores from individual modalities are ...

    最近文档搜索:

      站点地图 | 文档上传 | 侵权投诉 | 手机版
      本站所有资源均来自互联网,本站只负责收集和整理,均不承担任何法律责任,如有侵权等其它行为请联系我们.
      文档下载 Copyright 2013 doc.guandang.net All Rights Reserved.  -  email