本文从语音模型入手，讨论了新的抗噪语音识别方法。
This paper starts from the speech models, and main works focus on developing new noisy speech recognition methods.
因为这是没有必要来估计从嘈杂的语音模型的参数，非参数方法适合于更多的应用。
Because it is not necessary to estimate the model parameter from the noisy speech, the non-parameter methods are suitable for more applications.
无论如何它不是一个实际的文法—通过这么小的文法构建语音模型时，您很快就会遇到样例不够的问题。
It is not a practical grammar by any means-when building an acoustic model from grammars as small as this, you soon run into problems regarding insufficient samples.
各种改进算法针对语音模型、判决规则等核心问题进行了系统研究，检测性能得到了进一步提高。
Recent researches on speech model and decision rule improve the performance of voice activity detection.
针对特征语音说话人自适应算法的缺陷，提出了基于结构化特征语音模型的区别性说话人自适应方法。
A novel model-based speaker adaptation algorithm, support speaker weighting(SSW), was proposed for rapid speaker adaptation in speech recognition systems.
文中在一个大型数据库上对语音识别基元、语音模型、模型的输出观察向量的计分方法进行了大量的比较实验。
A great deal of experiments on speech recognition units, speech recognition models and the forms of scoring methods for output observation vectors have been done based on a giant speech corpus.
即使在模型精度不高的情况下，该方法也能得到合成质量比较好的语音。
The quality of the synthesized speech by this approach is rather high even though the model precision is low.
所建立的数学模型对自动语音导航系统的自主性深层开发，提供了有益的参考。
A beneficial reference is provided for deep independent development of automatic voice guide system with derived mathematic model.
支持向量机用作说话人确认模型来训练目标说话人和背景说话人的语音数据。
The support vector machine based speaker verification models are trained on the enrolled speaker and the background model.
该文研究了基于数据模拟方法和HMM（隐马尔科夫模型）自适应的电话信道条件下语音识别问题。
This paper addresses the problem of speech recognition under telephone channel conditions using data simulation method and HMM(Hidden Markov Model)adaptation.
波形内插语音编码模型作为当今最具潜力的低速率语音编码方案之一，因其良好的性能，越来越受到人们的重视。
As one of the current most potential low bit rate speech coding schemes, waveform interpolation model with its high performance has been paid more and more attention.
提出基于语音生成模型和发音模型RTLA合成模式实现以共振峰轨迹为目标的语音合成的新方法。
A new method to synthesize formant targeted sounds based on speech production model and RTLA articulatory synthesis model is studied.
该模型将呼叫的排队过程与语音包的排队过程合并起来，建立了完整的系统状态图。
Combine the queuing process of calling and the queuing process of voice packet the model set up an integrated system status diagram.
本文研究神经网络的多层感知器模型在语音识别中的应用。
This paper describes the use of multi-layer perception model of neural network in speech recognition.
最后，在中文语音识别任务上测试两类语言模型的性能。
Finally, these two language models are evaluated on the task of Chinese speech recognition.
本文应用全极点模型，提取语音信号的线性预测系数，并推导出其倒谱系数，获得线性预测倒谱差分，用以描述说话人声道的动态变化。
In this paper, we use full pole model to obtain speech signal LPC, then deduce it's LPCC, and we use the LPCC difference to describe speaker's track dynamic movement.
该模型将呼叫的排队过程与语音句的排队过程合并起来，建立了完整的系统状态图。
Combining the queuing process of calls and that of voice packets, the model sets up an integrated system status diagram.
新模型使语音识别率得到了改善。
本论文中，我们设计一个基于听觉感知模型的单耳语音分离系统。
In this thesis, we design such a monaural speech segregation system based on an auditory perceptual model.
提出了一个柔性可扩展体系结构非特定人语音识别系统的框架模型，介绍了相关实现原理。
A framework model of independent speech recognition system based on the flexible and extensible architecture is put forward, and some correlative theories are introduced.
通过应用全极点模型，提取语音信号的线性预测系数，并推导出其倒谱系数，获得线性预测倒谱差分，用以描述说话人声道的动态变化。
By using full pole model, we obtained speech signal LPC, then deduced it's LPCC, and we used the LPCC difference to describe speaker's track dynamic movement.
该方法采用了基于发音声学原理的反射型传输线模型来实现语音合成器。
The synthesis model is implemented with scattering process derived from reflection-type line analog of vocal tract system according to the acoustic model of speech production.
由于在语音识别中被广泛应用的隐马尔可夫模型(HMM)是一重马尔可夫模型，它不能充分地描述语音信号的时间相依性。
Since the widely used Hidden Markov model (HMM) in speech recognition is first order Markov model, it can not fully model the temporal dependence of speech signal.
双通路模型理论认为语音的通达存在两条通路，即词典通路和非词典通路。
There are two routes about phonological access in dual route model, lexical route and non-lexical route.
基于语音正弦模型，采用能量谱熵和子谐波谐波比率方法进行语音分析。
Energy-spectrum entropy and subharmonic-harmonic rate methods were used to analyze speech based on speech sinusoidal model.
测试实验证明了SDSPM模型在汉语语音识别中的有效性。
The SDSPM feasibility is tested in an actual Chinese speech recognition environment.
理论上详细介绍了HMM模型及与语音识别相关的语音数字信号处理。
This paper also introduces HMM model and speech digital signal processing associated with speech recognition .
讨论了一个用于语音编码的频域技术，用以在浊音语音共振峰模型方面改善低阶数线性预测(LP)的性能。
A frequency domain technique is presented to be used in speech coding to improve the performance of low order linear prediction (LP) in modeling of formants of voiced speech.
其中语音信号切割是经由语音辨认技术，依据声学模型的能量门限、图形比对的方法来切割出正确的发音区段。
Which voice signals cutting through voice recognition technology, based on the acoustic model energy threshold, graphics than a way to correct pronunciation sectors.
其中语音信号切割是经由语音辨认技术，依据声学模型的能量门限、图形比对的方法来切割出正确的发音区段。
Which voice signals cutting through voice recognition technology, based on the acoustic model energy threshold, graphics than a way to correct pronunciation sectors.
应用推荐