- Volume 20 Issue 8
In this paper, we propose the Recurrent Neural Predictive HMM (RNPHMM). The RNPHMM is the hybrid network of the recurrent neural network and HMM. The predictive recurrent neural network trained to predict the future vector based on several last feature vectors, and defined every state of HMM. This method uses the prediction value from the predictive recurrent neural network, which is dynamically changing due to the effects of the previous feature vectors instead of the stable average vectors. The models of the RNPHMM are Elman network prediction HMM and Jordan network prediction HMM. In the experiment, we compared the recognition abilities of the RNPHMM as we increased the state number, prediction order, and number of hidden nodes for the isolated digits. As a result of the experiments, Elman network prediction HMM and Jordan network prediction HMM have good recognition ability as 98.5% for test data, respectively.
- Technical Report ICS-8604 Serial Order: A parallel distributed processing approach M. I. Jordan
- Technical Report CRL-8801 Finding structure in time J. L. Elman
- A thesis for doctorate, Carnegie Mellon University Speech Recognition using Neural Networks J.Tebelskis
- 한국음향학회지 v.18 no.3 회귀신경망을 이용한 음성인식에 관한 연구 한학용;김주성;허강인
- Proc. int. conf. ASSP Combining hidden Markov model and neural network classifiers Nile L. T.;Silverman H. F.
- Proc. ICASSP'94 Vowel Classification Using A Neural Predicitive HMM: ADiscriminative Training Approach K. Hassanein;L. Deng;M. I. Elmasry
- the Journal of the Acoustical Society of Korea v.17 no.2E A Study on the Syliables Recognition Using Neural Network Predicitive HMM S. H. Kim;S. B. Kim;S. Y. Koh;K. I. Hur
- TENCON'99 v.Ⅰ A Study on the Recognition of the isolated Digits Using Recurrent Neural Predicitive HMM S. H. Kim;S. Y. Koh;K. I. Hur
- 確率モデルによる音聲認識 中川聖一
- 한국음향학회지 v.14 no.1 이산 지속시간제어 연속분포 HMM을 이용한 연속음성 인식 김수훈;이종진;허강인
- proc. ICASSP'90 Speaker-Independent Word Recognition Using A Neural Prediction Model Ken-Ichi Iso;Takao Watanabe