Description: 语音识别技术主要包括特征提取技术、模式匹配准则及模型训练技术三个方面。此外，还涉及到语音识别单元的选取。选择识别单元是语音识别研究的第一步。语音识别单元有单词（句）、音节和音素三种，具体选择哪一种，由具体的研究任务决定。-Speech recognition technology, including feature extraction techniques, pattern-matching criteria and the three aspects of model training techniques. In addition, it involves the selection of speech recognition modules. Choose identification unit is the first step in speech recognition research. Speech recognition module has the word (sentence), three types of syllables and phonemes, the specific selection of which, by the specific research mission decide. Platform: |
Size: 8192 |
Author:candy |
Hits:
Description: 先用端点检测将语音中有用的语音部分提取出来（即将头部和尾部的静音部分除掉），然后用LPC算法提取语音信号的特征参数，进行动态归整(DTW算法)后与模板库里面的标准语音作比较，最后将识别结果进行D/A转化后播放出来。在本部分的设计中，则主要完成语音识别的模式匹配算法部分的软件实现。 -First with the endpoint detection of speech to voice some of the useful extracted from the (soon to mute some of the head and tail removed), and then extract the speech signal with the LPC algorithm characteristic parameters of dynamic warping (DTW algorithm) after the template library compared to standard voice inside, and finally the recognition result is D/A conversion after playing out. In this part of the design, mainly to complete the pattern matching algorithm for speech recognition part of the software. Platform: |
Size: 6816768 |
Author:吴斌 |
Hits:
Description: 这是一段语音识别的c++源程序，包括预处理，端点检测，线性倒谱系数，dtw算法模式匹配。-This is a speech recognition c++ source, including preprocessing, detection, linear cepstrum, dtw algorithm for pattern matching. Platform: |
Size: 19456 |
Author:董芳 |
Hits:
Description: 这是一段关于语音处理与识别的程序，包括预处理，端点检测，线性倒谱系数求解，并运用dtw算法进行模式匹配。-This is a speech processing and recognition on the program, including pretreatment, endpoint detection, linear cepstrum solution, and the use of dtw algorithm for pattern matching. Platform: |
Size: 206848 |
Author:董小方 |
Hits:
Description: 按所学相关语音处理的知识，自己设计一个孤立词语音识别程序，分析所设计系统的特性。熟悉不同模块间如何进行有效的组合，以及模块内的设计，重点掌握特征参数的提取和模式识别方法，并对不同的特征参数提取方法和模式匹配方法有大概的了解，知道其不同的优缺点。-The learned voice processing knowledge, an isolated word speech recognition program of their own design, analysis of the design characteristics of the system. Familiar with the effective combination of how the different modules, and module design, focus on mastering the characteristic parameter extraction and pattern recognition methods, and different characteristic parameter extraction method and mode matching method probably know the different advantages and disadvantages . Platform: |
Size: 340992 |
Author:lin |
Hits:
Description: 语音特征信号识别是语音识别研究领域中的一个重要方面,一般采用模式匹配的原理解
决。语音识别的运算过程为:首先,待识别语音转化为电信号后输入识别系统,经过预处理后用数学方法提取语音特征信号,提取出的语音特征信号可以看成该段语音的模式。然后将该段语音模型同已知参考模式相比较,获得最佳匹配的参考模式为该段语音的识别结果.-Speech characteristic signal recognition is an important aspect in the field of speech recognition research, generally USES pattern matching the original understanding
Definitely. Operation process of speech recognition are as follows: first, after waiting for voice recognition into electrical signal input and recognition system, after pretreatment
Mathematical method was used to extract speech signal, and extracts the phonetic characteristics of signals can be as the segment of speech patterns. And then the
Audio model compared to a known reference model, get the best match the reference pattern for this speech recognition results. Platform: |
Size: 375808 |
Author:吴军 |
Hits:
Description: 语音特征信号识别是语音识别研究领域中的一个重要方面,一般采用模式匹配的原理解
决。语音识别的运算过程为:首先,待识别语音转化为电信号后输入识别系统,经过预处理后用数学方法提取语音特征信号,提取出的语音特征信号可以看成该段语音的模式。然后将该段语音模型同已知参考模式相比较,获得最佳匹配的参考模式为该段语音的识别结果-Recognition is the speech characteristic signal in the field of speech recognition is an important aspect, the general principle of pattern matching solution. Voice recognition operation process as follows: First, to be converted to electrical signals identifying the input speech recognition system, after pretreatment using mathematical methods to extract the speech characteristic signal, the speech characteristic signal extracted can be seen that the pattern of voice. Then the paragraph speech model with a known reference pattern compared to the best matching reference pattern of speech recognition results for the segment Platform: |
Size: 489472 |
Author:吴军 |
Hits:
Description: mfcc特征提取， dtw匹配，老外写的，值得学习。- feature extraction, including MFCC and HFCC features, widely used in
speech recognition
* a simple DTW (dynamic time warping) pattern matching algorithm which
also can be used for recognition
Platform: |
Size: 250880 |
Author:yangf |
Hits:
Search - speech pattern matching