feature of speech
英
美
[网络] 演讲特点
双语例句
- MFCC feature extraction of speech based on pitch period
基于基音周期的语音MFCC参数提取 - And then introduces the functions and key technologies of pre-processing 、 feature extraction pattern matching and post-processing of speech recognition. Improved methods have been proposed in view of problems existed in traditional methods.
然后分别介绍了语音识别的预处理、特征参数提取、模式匹配和后处理阶段的功能及其关键技术，并针对传统方法中存在的问题提出了改进方案。 - Design and analysis of acoustic feature for corpus of speech synthesis
语音合成语料库的设计与声学特征分析 - The colloquialism is the most distinctive feature of the host's verbal speech and the host's gestures, which should conform to the plain, authentic, natural and friendly context, are closely associated with the host's refinement and creativity.
播报的口语化是电视新闻节目主持人有声语体构成的最大特色，主持人的体态语体必须符合新闻节目朴素、真实、自然、亲切的语境，同时与主持人的内在修养和创造性密切相关。 - The system performance of dialect identification depends on feature extraction of the speech signal, and the reasonable selection of characteristic parameters can greatly improve the recognition rate on dialect identification system.
方言辨识系统性能好坏取决于语音信号特征的提取，合理地选择特征参数对方言辨识系统的识别率有很大的提高。 - The problem of syntactic feature of speech is one of the important contents in syntactic study.
词类的句法特征问题是句法研究中的重要内容之一。 - First of all, I analyse the prosodic feature of the news broadcasting speech with the knowledge of acoustic-phonetics. For example, the extending of the syllable duration is the acoustic characteristics of the pre-boundary syllables and the accented syllables.
首先，通过实验的手段分析了新闻播音语言韵律特征的语音声学表现：即音节时长的拉长是边界前音节和重音音节的声学征兆； - This paper uses wavelet theory in noise-robust feature extraction of speech recognition and introduces a feature extraction method based on Gauss wavelet filter. The Gauss wavelet filter with human critical frequency band is obtained by studying human auditory characteristics.
把小波理论应用于抗噪语音识别特征提取，提出了基于高斯小波滤波器的语音识别特征提取方法，通过对人耳听觉特性的研究，按照人耳临界带宽设计了一组高斯小波带通滤波器。 - Finally, according to the feature of speech recognition system which is designed in the paper, the principle of DTW arithmetic is discussed to be the recognition method of this system and the simulation process of the recognition voice is done using this method.
根据本文设计的语音识别系统的特点，确定了动态时间归正（DTW）语音识别方法作为本系统的识别方法，并采用该算法对要识别的语音进行了仿真，得到正确的识别结果。 - During simulation experiment, wavelet analysis technique is adopted to extract feature vectors of speech, the results show that SVM and FSVM have both higher correct recognition rate and shorter training time than RBF network.
在仿真实验中，采用小波分析方法提取语音特征向量，识别结果表明，SVM和FSVM比RBF网络具有较好的泛化性能，训练时间也大大缩减。