的连续语音识别直到最近才变得实用。
Only recently, for example, has continuous speech recognition for dictation become available.
系统中的语音识别引擎使用的是大词汇量连续语音识别技术。
The system employs large vocabulary continuous speech recognition engine in front speech input end.
本文介绍了在连续语音识别和语音机器翻译方面所进行的工作。
Continuous speech recognition and automatic speech translation are important subjects. In this paper, our results on these topics are introduced.
基于特征口形的音频-视频混合连续语音识别系统具有很好的抗噪性。
The experiments also show that the audio-visual continuous speech recognition system is robust in noisy environments.
基于特征口形的音频-视频混合连续语音识别系统具有很好的抗噪性。
The experiments also show that the audio-visual continuous speech recognition system is…
因此带噪环境的连续语音识别一直是当前语音信号研究的热点和难点。
So continuous speech recognition in noisy environment is always the hot spot and the difficult spot of the speech signal research.
大词表连续语音识别系统由多个组件构成，识别错误受多种因素的影响。
Large vocabulary continuous speech recognition system consists of several components, and recognition errors are caused by different factors.
本文研究了大词汇量非特定人汉语连续语音识别和理解系统中的容错技术。
In this paper, error tolerant techniques are studied for large vocabulary speaker independent Chinese continuous speech recognition and understanding systems.
识别实验表明，文中建立的音频视频连续语音识别系统具有很好的抗噪性。
This paper presents an audiovisual continuous speech recognition system for noise environments.
目前，汉语连续语音识别在纯净语音环境下，已经取得了较好的识别效果。
At present, Chinese continuous speech recognition has got a good performance in pure speech environment.
而语言模型、语法及词法模型在中、大词汇量连续语音识别中是非常重要的。
Furthermore it is very important that we use language model, syntax and accidence model in middle or big glossary continuous speech recognition.
本课题旨在研究和开发一个具有自然交谈风格的大词汇量连续语音识别系统。
The research will lead to a new recognition system for the spontaneous and conversational speech.
该系统利用中大词汇量非特定人连续语音识别技术实时实现了邮包信息的语音校核。
This system can achieve the parcel post information checking and verification by means of a medium vocabulary speaker independent continuous speech technology.
汉语普通话是一种带声调的语言，声调信息在汉语连续语音识别中具有非常重要的作用。
Chinese is a langue with tone, and the tone information is very important for the recognition of Chinese speech.
提出一种模式识别方法，它能有效地用于有监督的连续语音识别、文稿理解和数据文件分析。
This paper raised a method of pattern recognition, it can be used to the continuous speech recognition with reference, text understanding and data file analysis.
大词汇连续语音识别是当前语音识别研究的主流，探讨其中声调识别的特殊性是一项有意义的工作。
Large vocabulary continuous speech recognition is main direction of speech recognition research at present, in which the research on particularity of tone recognition is an important work.
该系统利用大词汇量非特定人连续语音识别技术与口语对话模型实现了智能熊猫系统的人机知识问答。
Human-ma-chine interlocution of Smartpanda System is implemented by means of speaker independent continuous speech recognition technology and dialogue model.
本文在简单介绍词组语音识别之后给出了一个基于连续语音识别算法和词树约束的汉语词组语音识别方法。
After an overview of phrase recognition, this paper proposes a Chinese phrase recognition method based on continuous speech recognition algorithm and phrase-tree constraint.
对大词汇量汉语连续语音识别的实验结果表明：高斯模糊聚类使高斯数减少25%时，识别率提高了0.15%。
The experimental results on large vocabulary continuous Mandarin speech recognition show when the number of Gaussians is reduced by 25%, the recognition accuracy increases by 0.15%.
文章分别对连续语音识别中可能出现的关联词技术、关键词技术、顽健性技术、自适应性技术及搜索策略给予了介绍。
In this paper, it respectively introduces the connection words technique, keyword technique, persistent technique, adaptation technique, searching strategy possibly in continual speech recognition.
动态贝叶斯网络(DBN)，以其扩展性和对时间序列的强大描述、推导和学习能力，逐渐被应用于连续语音识别中。
Dynamic Bayesian Network (DBN), because of extensibility, powerful description, inference and learning abilities for the time series, being used in the speech recognition.
他曾开发出世界上第一个“非特定人连续语音识别系统”，并被《商业周刊》1988年授予 “最重要科学创新奖”。
The first "Speaker-Independent Continuous Speech Recognition" system in the world developed by him was awarded the "Most Important Innovation of 1988" by Business Week magazine.
完整的连续语音识别系统主要包括四个部分：特征提取，声学模型，语言模型和搜索算法，本文就是根据这四个部分展开的。
A whole continuous speech recognition system includes four parts: feature extraction, acoustic model, language models and search algorithms, and the thesis is carried out according to them.
对于特定人连续语音识别系统和非特定人连接词识别系统，在使用不同的特征矢量时，不同编码器产生的识别效果有比较大的差异。
Using different feature vectors, the effect of the vocoders differs on speaker dependent continuous speech recognition and speaker independent connected digits recognition.
最后，完成了基于本文混合模型的非特定人小词汇量连续语音识别系统的设计和实现，通过大量试验验证了此混合模型的有效性和适用性。
Finally, it design a small-vocabulary continuous speech recognition system based on the mixed model, through experimental verify the validity and serviceability of the mixed model.
三字词音节声调模式具有连续语音中音节声调模式的特征，声调的提取和识别远较孤立字困难。
Tone extraction and recognition of trisyllabic word is more difficult than that of isolated word, because it holds the characteristic of syllable in continuous speech.
同时，根据语音信号不连续的特点，压缩算法具有静音识别功能，进一步提高了语音信号的压缩率。
According to the characteristic of discontinuous voice signal, this compression algorithm provides the function of voice active detection and improves the compress ratio of voice signal.
若能把连续大词汇量的语音进行实时自动切分为单个音节，便可提高系统的识别率。
If speech sound of continuous large vocabulary can be automatically divided into segments of single syllable, the recognition ratio of system can be raised.
将基于连续隐含Markov模型语音 识别算法中占系统总运算量的50%以上的Mahalanobis距离 计算，映射为硬件实现的 模块。
The mapped algorithm, called the Mahalanobis distance, handles about 50% of its computational load in the overall speech recognition algorithm using a continuous hidden Markov model(CHMM).
实验表明，最终汉语连续数字语音识别率有很大提高。
Experiments show that the proposed algorithm can improve the connected digital speech recognition performance.
应用推荐