| Peripheral Nonlinear Time Spectrum Features Algorithm for Large Vocabulary Mandarin Automatic Speech Recognition |
| |
| 作者姓名： | Fadhil H.T.Al-dulaimy 王作英 |
| |
| 作者单位： | Department of Electronic Engineering，Tsinghua University，Beijing 100084，China，Department of Electronic Engineering，Tsinghua University，Beijing 100084，China |
| |
| 基金项目： | Supported by the National High-Tech Research and Development (863) Program of China (No. 200/AA/14) |
| |
| 摘 要： | IntroductionCurrentautomaticspeechrecognitionsystemsarebasedoncontext-dependentorcontext-independentphonicsorsyllablemodelsdescribedintermsofse-quencesofhiddenMarkovmodel(HMM)states,whereeachstateisassumedtobecharacterizedbyastationaryprobabilitydensityfunction.Thetimecorre-lationandconsequently,thesignaldynamicsinsideeachHMMstate,arealsousuallydisregardedalthoughtheuseofdynamicfeatures,suchasdeltaanddelta-deltaparameters,cancapturesomeofthecorrelations.Consequently,onlymedium-termdependenc…
|
| 关 键 词： | 非线性时间光谱特征 计算方法 自动语音识别系统 词汇 |
| 收稿时间： | 1 September 2003 |
Peripheral Nonlinear Time Spectrum Features Algorithm for Large Vocabulary Mandarin Automatic Speech Recognition |
| |
| Fadhil H. T. Al-dulaimy,WANG Zuoying.Peripheral Nonlinear Time Spectrum Features Algorithm for Large Vocabulary Mandarin Automatic Speech Recognition[J].Tsinghua Science and Technology,2005,10(2):174-182. |
| |
| Authors: | Fadhil H T Al-dulaimy WANG Zuoying |
| |
| Institution: | Fadhil H. T. Al-dulaimy,WANG Zuoying Department of Electronic Engineering,Tsinghua University,Beijing 100084,China |
| |
| Abstract: | This work describes an improved feature extractor algorithm to extract the peripheral features of point x(ti,fj) using a nonlinear algorithm to compute the nonlinear time spectrum (NL-TS) pattern. The algorithm observes n×n neighborhoods of the point in all directions, and then incorporates the peripheral features using the Mel frequency cepstrum components (MFCCs)-based feature extractor of the Tsinghua electronic engineering speech processing (THEESP) for Mandarin automatic speech recognition (MASR) system as replacements of the dynamic features with different feature combinations. In this algorithm, the orthogonal bases are extracted directly from the speech data using discrite cosime transformation (DCT) with 3×3 blocks on an NL-TS pattern as the peripheral features. The new primal bases are then selected and simplified in the form of the Δdp-t operator in the time direction and the Δdp-f operator in the frequency direction. The algorithm has 23.29% improvements of the relative error rate in comparison with the standard MFCC feature-set and the dynamic features in tests using THEESP with the duration distribution-based hidden Markov model (DDBHMM) based on MASR system. |
| |
| Keywords: | large vocabulary speech recognition Mandarin automatic speech recognition (MASR) dura- tion distribution-based hidden Markov model (DDBHMM) feature identification |
| 本文献已被 CNKI 万方数据 ScienceDirect 等数据库收录！ |
