speech database

Data Model and Implementation Technology of a Speech Database

语音数据库的数据模型及实现技术

It is building up its database of speech during Siri's early life.

在Siri的早期阶段，它的语音数据库逐渐建立。

Architecture of fast speech database is also presented.

提出了快速数据库架构。

In order to realize visual speech synthesis, a satisfied bimodal database needs to be built up.

为实现可视语音合成, 建立符合条件的双模态语料库,提出双模态语料自动选取算法.

Spectral envelope conversion is an important content of voice conversion. Based on bilinear transform function, spectral envelope can be efficiently converted even with small training speech database.

频谱包络转换是语音转换中的一项重要内容，基于双线性转换函数进行频谱搬移的方法可以有效的进行谱包络转换，并且基于小语料库训练时仍能得到稳定的转换谱包络。

Secondly, the experimental Chinese emotional speech database is recorded.

其次为实验录制了汉语情感语音库。

A Decision Tree Based Approach for Construction of Speech Database and Unit Selection

一种基于决策树模型的音库构建和基元选取方法

The training and testing sample images are from Chinese Audio Visual Speech Recognition Database ( CAVSR v1.0).

该文实验中所使用的样本数据来自中国科学院声学所汉语听觉、视觉双模态数据库（CAVSRv1.0）。

The building of Speech Database of Janet System and exploring for a special Chinese speech phoneme

知音文书机语音数据库的建立和一种汉语语音特性的探讨

Automatic Text Selection for Continuous Speech Database of Standard Chinese Based on Comprehensive Factors

基于综合因素的汉语连续语音库语料自动选取

Design of speech database for Chinese names based on speech synthesis

基于语音合成法的汉语人名语音库的设计应用

The method of speech synthesize can improve naturalness of speech basing on the database of speech.

基于语音数据库的语音合成方法有望进一步提高语音合成的自然度.

Using Wavelet Transform to detect the Glottal Closure Instant ( GCI), proposed pitch-synchronization and sample selection LP analysis based on the GCI information, the obtained parameters would be saved in the speech database.

利用小波变换检测语音信号的声门闭合时刻（GCI），利用语音信号的GCI进行基音同步和样本选择的多脉冲线性预测分析，将得到的语音参数保存到语音库中。

In Discrete Hidden Markov Model ( DHMM) isolated word recognition system, we use adaptation method based on hierarchical spectral clustering. We also use new speech Database containing 1000 people for Speaker-Independent speech recognition.

在离散隐马尔可夫模型DHMM孤立字语音识别系统中，采用了分级谱分类的说话人自适应方法，并将新建的1000人的语音库用于不定人识别。

The study of declination in speech database in standard Chinese

基于言语数据库的汉语音高下倾现象研究

Has introduced the connect of surd and sonant detailedly. ( 3) With the algorithm in this paper, we have accomplished speech name database of intelligence telephone-based military recruiting system.

采用本文算法，建立了军队智能电话招生信息系统的语音姓名库。