Description: 说话人识别是语音识别的一种特殊方式，其目的不是识别语音内容，而是识别说话人是谁，即从语音信号中提取个人特征。采用矢量量化（VQ）可避免困难的语音分段问题和时间归整问题，且作为一种数据压缩手段可大大减少系统所需的数据存储量。本文提出了识别特征选取采用复倒谱特征参数和对应用VQ的说话人识别系统改进的一种方法。当用于训练的数据量较小时，复倒谱特征可以得到比较稳定的识别性能。VQ的改进方法避免了说话人识别系统的训练时间与使用时间相差过长从而导致系统的性能明显下降以及若利用自相关函数带来的大量运算。-Speaker Recognition Speech Recognition is a special way, its purpose is not voice recognition, Who identification but said that the voice signal from extracting personal characteristics. Vector quantization (VQ) can avoid the difficulties subparagraph voice to the issues and the whole time, and as a means of data compression system can significantly reduce the required data storage capacity. This paper presents a selection of identifiers employ Cepstrum parameters and the application of VQ speaker recognition system to improve a side France. When training for the amount of data is smaller, rehabilitation Cepstrum be relatively stable recognition performance. VQ improved ways to avoid the speech recognition system of training and the use of the difference in time, resulting in excessive sys Platform: |
Size: 23925 |
Author:张开 |
Hits:
Description: 说话人识别是语音识别的一种特殊方式，其目的不是识别语音内容，而是识别说话人是谁，即从语音信号中提取个人特征。采用矢量量化（VQ）可避免困难的语音分段问题和时间归整问题，且作为一种数据压缩手段可大大减少系统所需的数据存储量。本文提出了识别特征选取采用复倒谱特征参数和对应用VQ的说话人识别系统改进的一种方法。当用于训练的数据量较小时，复倒谱特征可以得到比较稳定的识别性能。VQ的改进方法避免了说话人识别系统的训练时间与使用时间相差过长从而导致系统的性能明显下降以及若利用自相关函数带来的大量运算。-Speaker Recognition Speech Recognition is a special way, its purpose is not voice recognition, Who identification but said that the voice signal from extracting personal characteristics. Vector quantization (VQ) can avoid the difficulties subparagraph voice to the issues and the whole time, and as a means of data compression system can significantly reduce the required data storage capacity. This paper presents a selection of identifiers employ Cepstrum parameters and the application of VQ speaker recognition system to improve a side France. When training for the amount of data is smaller, rehabilitation Cepstrum be relatively stable recognition performance. VQ improved ways to avoid the speech recognition system of training and the use of the difference in time, resulting in excessive sys Platform: |
Size: 23552 |
Author:张开 |
Hits:
Description: unix下的语音信号处理工具库，用c语言编写，可用于lsp、lpc、矢量量化分析等等。-under unix tools for speech signal processing library, with c languages, can be used for lsp, lpc, analysis and so on vector quantization. Platform: |
Size: 1105920 |
Author:李航 |
Hits:
Description: 对于语音编码参数的有权重多级矢量量化，主要是对LSP码本训练，这个权重学习语音的都应知道，且这个可以训练大的语音参数，完全自已写的。上传前已调试成功。-Speech coding parameters for the right to re-multistage vector quantization, mainly for LSP codebook training, the right to re-learn voice should know, and that this can be trained big voice parameters, completely their own writing. From the success of pre-commissioning. Platform: |
Size: 6682624 |
Author:马庆利 |
Hits:
Description: 采用矢量量化对语音信号量化，并与标量量化对比分析，绘制了语音信号的概率密度曲线-The use of vector quantization for speech signal quantization and scalar quantization and comparative analysis, rendering the speech signal probability density curve Platform: |
Size: 160768 |
Author:wu wei |
Hits:
Description: 本文首先介绍了目前语音识别的发展现状和主要手段，分析了语音识别中所采用的主要特征参数和比较前沿的研究方向，另外着重讲解了语音识别中最常用隐马尔可夫H(MM)模型，及应用广泛的矢量量化方法(VQ)。接着介绍了嵌入式平台，从软、硬件方面着重介绍了与语音识别相关部分的设计包括硬件及相关驱动程序设计，最后介绍了系统实现方法与测试结果。
-This paper describes the current status of the development of speech recognition and the main means of analysis used in speech recognition parameters and compare the main features cutting-edge research, while focused on explaining the most commonly used in speech recognition, Hidden Markov H (MM) model of , and the widely used vector quantization (VQ). Then introduced the embedded platform, from hardware and software aspects of highlighting the relevant parts of speech recognition with the design, including hardware and associated drivers for the design, finally introduced a system implementation method and test results. Platform: |
Size: 6030336 |
Author:fff |
Hits:
Description: This paper presents results of speaker recognition experiments using short Polish sentences. We developed and analyzed various vector quantization representations in order to first maximize identification effectiveness and second to compare VQ (vector quantization) and GMM (Gaussian mixture model) approaches. For the research and experiments we created and exploited database, containing specially prepared short speech sequences. Platform: |
Size: 258048 |
Author:Tomasz |
Hits:
Description: ector quantization is a classical quantization technique from signal processing which allows the modeling of probability density functions by the distribution of prototype vectors. It was originally used for data compression. It works by dividing a large set of points (vectors) into groups having approximately the same number of points closest to them. Each group is represented by its centroid point, as in k-means and some other clustering algorithms.
Digital libraries not only consist of text data, but also speech and image data. To compress speech data techniques such as vector quantization (VQ) are used. Platform: |
Size: 6144 |
Author:noopur |
Hits:
Description: SPTK是一套语音信号处理工具为UNIX环境中,如LPC分析,PARCOR分析、LSP分析,PARCOR合成过滤器,LSP合成过滤器,矢量量化技术,以及其他扩展的版本。-SPTK is a suite of speech signal processing tools for UNIX environments, e.g., LPC analysis, PARCOR analysis, LSP analysis, PARCOR synthesis filter, LSP synthesis filter, vector quantization techniques, and other extended versions of them. Platform: |
Size: 6351872 |
Author:石中华 |
Hits:
Description: LBG分类算法
用初始室心随机法和扰动因子分裂法两种方法，比较不同方法不同参数设置时的分类性能。
-LBG classification algorithm vector quantization: vector normalization within a certain range for a particular type, consists of two steps: first generate a codebook, which is the speech feature vector space by the first process- also known as clustering speech parameter sequence as a vector, the reference code for classified- also known as quantization. Clustering algorithm: it is relatively simple and commonly used K-means clustering algorithm. LBG is a clustering algorithm, which is generally assumed that the codebook size is fixed, and for a power of 2. Codebook is small, then expanding until it reaches the requirements. It is often an existing classification split into two subclasses, and initial value with the new code word to each subclass. LBG algorithm on random data and a certain regularity (and meet certain Gaussian distribution) data classification, and look at the performance of the LBG algorithm, the initial chamber heart random disturbance factor-secession law are two Platform: |
Size: 86016 |
Author:zzc |
Hits:
Description: 应用LBG算法对语音信号进行矢量量化。本压缩包共有两个主文件，training.m对训练数据进进行处理并的到四个初始码本，quantizing.m对待量化数据进行矢量量化。其余为自编功能函数。
-The application of the LBG algorithm for vector quantization of speech signals. The compressed package a total of two main documents training.m into the training data to be processed and to the four initial codebook, quantizing.m treatment of quantitative data for vector quantization. The rest is self performance function. Platform: |
Size: 4096 |
Author:fz |
Hits:
Description: 用于聚类的矢量量化程序，可用在相关领域，程序本为语音识别设计，可正常运行。-Used for clustering vector quantization procedure, can be used in the related field, program designed for speech recognition, can be normal operation
Platform: |
Size: 1551360 |
Author:vsdlh ausdhf |
Hits:
Description: 针对语音信号的能量，实现三维矢量量化，量化失真小。-Energy for the speech signal to achieve a three-dimensional vector quantization, a quantization distortion. Platform: |
Size: 342016 |
Author:hy |
Hits:
Description: In this work, the Mel frequency Cepstrum Coefficient
(MFCC) feature has been used for designing a text
dependent speaker identification system. The extracted
speech features (MFCC’s) of a speaker are quantized to a
number of centroids using vector quantization algorithm.
These centroids constitute the codebook of that speaker.
MFCC’s are calculated in training phase and again in testing
phase. Speakers uttered same words once in a training
session and once in a testing session later. The Euclidean
distance between the MFCC’s of each speaker in training
phase to the centroids of individual speaker in testing phase
is measured and the speaker is identified according to the
minimum Euclidean distance. The code is developed in the
MATLAB environment and performs the identification
satisfactorily. Platform: |
Size: 1024 |
Author:abdou |
Hits:
Search - vector quantization for speech