Speech recognition with hierarchical recurrent neural networks

Pattern Recognition

Volumn 28, Issue 6, 1995, Pages 795-805

  Chen, Wen Yuan ^a   Liao, Yuan Fu ^a   Chen, Sin Horng ^a  

^a NATIONAL CHIAO TUNG UNIVERSITY   (Taiwan)

Author keywords

Discriminative training; Generalized probabilistic descent; Hierarchical; Recurrent neural networks; Speech recognition

Indexed keywords

ALGORITHMS; HIERARCHICAL SYSTEMS; MARKOV PROCESSES; MATHEMATICAL MODELS; NEURAL NETWORKS; PROBABILITY;

DISCRIMINATIVE TRAINING; GENERALIZED PROBABILISTIC DESCENT (GPD) ALGORITHM; HIERARCHICAL RECURRENT NEURAL NETWORK (HRNN);

SPEECH RECOGNITION;

EID: 0029327528     PISSN: 00313203     EISSN: None     Source Type: Journal    
DOI: 10.1016/0031-3203(94)00145-C     Document Type: Article

References (24)

1. Time warping recurrent neural networks
2. Word recognition based on the combination of a sequential neural network and the GPDM discriminative training algorithm
3. A time-delay neural network architecture for isolated word recognition
4. Speaker-independent word recognition using dynamic programming neural networks
5. Speech recognition using dynamical model of speech production
6. Performance through consistency connectionist large vocabulary continuous speech recognition
7. Dynamic programming algorithm optimization for spoken word recognition
8. Real time Chinese syllable recognition with hierarchically structures neural network and transputer system
9. Multi-speaker/speaker independent architectures for the multi-state time delay neural network
10. The meta-pi network: connectionist rapid adaption for high performance multi-speaker recognition
11. The meta-pi network building distributed knowledge representations for robust multisource pattern recognition
12. Syllable recognition using integrated neural networks
13. Modularity and scaling in large phonemic neural networks
14. Initial experiments on constructing domains of expertise and hierarchies in connectionist systems
15. Dynamic spectral shape features as acoustic correlates for initial stop consonants
16. Application of fully recurrent neural networks for speech recognition
17. Connected recognition with a recurrent network
18. New discriminative training algorithms based on the generalized probabilistic descent method
19. Discriminative training of dynamic programming based speech recognizes
20. Discriminative training
21. GPD training of dynamic programming-based speech recognizers
22. Learning internal representation by error propagation
23. Vector quantization of pitch information in Mandarin speech
24. The Speech database of Telecommunication Laboratories, Ministry of Transportation and Communications, R.O.C.