Duration Modeling Using Cumulative Duration Probability
スポンサーリンク
概要
- 論文の詳細を見る
A duration modeling technique is proposed for the HMM based connected digit recognizer. The proposed duration modeling technique uses a cumulative duration probability. The cumulative duration probability is defined as the partial sum of the duration probabilities which can be estimated from the training speech data. Two approaches of using it are presented. First, the cumulative duration probability is used as a weighting factor to the state transition probability of HMM. Second, it replaces the conventional state transition probability. In both approaches, the cumulative duration probability is combined directly to the Viterbi decoding procedure. A modified Viterbi decoding procedure is also presented. One of the advantages of the proposed duration modeling technique is that the cumulative duration probability rules the transitions of states and words at each frame. Therefore, an additional post-procedure is not required. The proposed technique was examined by recognition experiments on Korean connected digit. Experimental results showed that two approach achieved almost same performances and that the average recognition accuracy was enhanced from 83.60% to 93.12%.
- 社団法人電子情報通信学会の論文
- 2002-09-01
著者
-
Youn Dae-hee
The Dept. Of Electrical And Electronic Eng. Yonsei Univ. Seoul Korea.
-
Youn D‐h
Yonsei Univ. Seoul Kor
-
Youn Dae-hee
Department Of Electrical And Electronic Engineering Yonsei University
-
Youn Dae-hee
The Department Of Electrical And Electronic Eng. Yonsei Univ.
-
YANG Tae-Young
the LG Electronics Institute of Technology, Seoul, Korea.
-
LEE Chungyong
the Dept. of Electrical and Electronic Eng., Yonsei Univ., Seoul, Korea.
-
Lee Chungyong
The Dept. Of Electrical And Electronic Eng. Yonsei Univ. Seoul Korea.
-
Lee Chungyong
The Department Of Electrical And Electronic Eng. Yonsei Univ.
-
Yang Tae-young
The Lg Electronics Institute Of Technology Seoul Korea.
関連論文
- Bark Coherence Function for Speech Quality Evaluation over CDMA System
- An Efficient Active Noise Control Algorithm Based on the Lattice-Transversal Joint(LTJ) Filter Structure
- A GMM-Based Feature Selection Algorithm for Multi-Class Classification
- Design of Time-Varying Reverberators for Low Memory Applications
- Duration Modeling Using Cumulative Duration Probability
- A Grammatical Structure of the FSN for the Recognition of Korean Price Sentences
- Speaker Adaptation Based on a Maximum Observation Probability Criterion
- A Very Low Complexity VSELP Speech Coder Using Regular Pulse Basis Vectors (Special Section of Papers Selected from ITC-CSCC'96)
- Performance Comparison of Single and Multi-Stage Algebraic Codebooks(Speech and Hearing)
- Performance Comparison of Single and Multi-Stage Algebraic Codebooks
- A New Method for Calibration of NLOS Error in Positioning Systems
- Voice Conversion Using Low Dimensional Vector Mapping(Regular Section)