During the procedure of speech recognition to the chinese continuous speech segments , there always brings the decreasing of the recognition correctness rate due to the wrong estimation for the start and the end of a piece of speech segment 在汉语连续语音识别过程中，经常对语音端点的判断发生错误，导致了识别成功率的下降。
In this dissertation tone patterns of disyllabic word and isolated - word are introduced , tone patterns for tn - syllabic word are analyzed in detail . tone patterns for continuous speech are researched based on the foregoing work 本文介绍了孤立字和二字词的声调模式，定性和定量地分析了三字词的声调模式，在此基础上研究了连续语音的声调模式。
Comparing with the continuous speech recognition , keyword spotting has advantages of less time consuming , higher calculation and stronger practicability and so on . thus this technology has practical application in many areas 相对于连续语音识别来说,关键词检出技术有资源与时间耗费少,正确率高,实用性强等特点,所以它在很多领域有着良好的应用前景。
In this thesis , first we analyzed and designed a traditional continued speech recognition system , which based on hmm and mfcc speech features . then we researched some noise robust technologies based on that system 本论文首先分析并实现了一个以mel频率倒谱系数( mfcc )作为语音特征,基于隐马尔可夫模型( hmm ) ,针对连续数字串识别任务的基本连续语音识别系统。
Viberbi beam search algorithm based on dynamic programming has become increasingly popular in csr , with which knowledge sources can be efficiently organized to constrain search space and different pruning strategies can be easily combined 为了有效地组织各种知识源，约束搜索空间，基于动态规划思想的viterbibeam搜索算法成为目前连续语音识别主要采用的方法。
Tone characteristics of syllable in continuous speech have lager variation than original tone characteristic under the influence of its preceding and posterior syllable , so tone patterns is more complicated , and have only basic characteristic of chinese tone 合肥工业大学搏土论文扬耍（ 4 ）声调模式分析。连续语音中各音节的声调特征受前后音节的影响变化较大，声调模式更加复杂，仅具有四声的基本特征。
Although there still exist many problems which need to be overcome such as the technique of large - vocabulary , speaker - independent , continuous speech recognition , the technique ( lpc and dtw ) for small - vocabulary , speaker - dependent , isolated - word speech recognition has come to the age of maturity 虽然大字表非特定人连续语音识别技术仍有许多难点尚未解决,但特定人小字表孤立词语音识别技术( lpc和dtw算法)已相对成熟。
Tone information isn ? basically used in current chinese speech recognition system , study on tone recognition is limited to tone recognition of isolated - word and multi - syllabic word , and research on tone patterns and tone recognition for chinese continuous speech is little 现有的汉语语音识别系统基本上没有利用声调信息，声调识别研究也多限于孤立字和多字词的声调识别，连续语音的声调模式和声调识别研究很少，本文在这方面开展了一点工作。
Because noise causes the mismatch between the acoustic models and the testing speech , the performance of speech recognition systems will degrade rapidly in noisy environments . therefore , noise robust technology is a very crucial problem for the speech recognition 目前的连续语音识别系统对纯净语音已能达到非常高的识别精度,但是无处不在的背景噪声带来了训练模型和测试语音之间的失配,这种失配使得连续语音识别系统的性能在噪声环境中急剧下降。
According to calculate lots of speech segments , because the author applies the new recognition characters to the recognition procedure and selects the appropriate parameter values , the author ' s new method improves the recognition correctness rate for the start and the end of a piece of chinese continuous speech segment 根据对大量实验语音的统计结果，由于使用了新的识别特征，并且选择了合适的重要参数值，笔者的方法提高了汉语连续语音端点检测的识别正确率。
