A modular RNN-based method for continuous Mandarin speech recognition

IEEE Transactions on Speech and Audio Processing

Volumn 9, Issue 3, 2001, Pages 252-263

  Liao, Y F ^a   Chen, S H ^a  

^a NATIONAL CHIAO TUNG UNIVERSITY   (Taiwan)

Author keywords

Mandarin speech recognition; MCE GPD algorithms; Modular recurrent neural networks

Indexed keywords

ALGORITHMS; DYNAMIC PROGRAMMING; LEARNING SYSTEMS; MARKOV PROCESSES; MATHEMATICAL MODELS; PROBABILITY; RECURRENT NEURAL NETWORKS; SPEECH ANALYSIS;

CONTINUOUS MANDARIN SPEECH RECOGNITION; DYNAMIC PROGRAMMING BASED RECOGNITION SEARCH MODULE; GENERALIZED PROBABILISTIC DESCENT; MINIMUM CLASSIFICATION ERROR;

CONTINUOUS SPEECH RECOGNITION;

EID: 0035274629     PISSN: 10636676     EISSN: None     Source Type: Journal    
DOI: 10.1109/89.905999     Document Type: Article

References (38)

1. A tutorial on hidden Markov models and selected applications in speech recognition
2. Continuous speech recognition - an introduction to the hybrid HMM/connectionist approach
3. Connectionist Viterbi training: A new hybrid method for continuous speech recognition
4. An application of recurrent nets to phone probability estimation
5. Task decomposition through competition in a modular connectionist architecture: The what and where vision tasks
6. Design and evolution of modular neural network architectures
7. Modularity and scaling in large phonemic neural networks
8. The Meta-Pi network: Building distributed knowledge representations for robust multi-source pattern recognition
9. Hierarchical mixtures of experts and the EM algorithm
10. ACID/HNN: Clustering hierarchies of neural networks for context-dependent connectionist acoustic modeling
11. Modular recurrent neural networks for Mandarin syllables recognition
12. A survey on automatic speech recognition with an illustrative example on continuous speech recognition of Mandarin
13. Continuous Mandarin speech recognition for Chinese language with large vocabulary based on segmental probability model
14. Modeling context-dependent phonetic units in a continuous speech recognition system for Mandarin Chinese
15. Context-dependent acoustic models for Chinese speech recognition
16. Phonetic modeling in the Philips Chinese continuous-speech recognition system
17. Improved syllable-based continuous Mandarin speech recognition using inter-syllable boundary models
18. An RNN-based preclassification method for fast continuous Mandarin speech recognition
19. Statistical and discriminative methods for speech recognition
20. Minimum classification error rate methods for speech recognition
21. An N-best candidates-based discriminative training for speech recognition application
22. Robust parametric modeling of durations in hidden Markov models
23. Forward-backward retraining of recurrent neural networks
24. Context-dependent classes in a hybrid recurrent network-HMM speech recognition system
25. Golden Mandarin (I) - a real-time Mandarin speech dictation machine for Chinese language with very large vocabulary
26. Complete recognition of continuous Mandarin speech for Chinese language with very large vocabulary using limited training data
27. Discriminative training of dynamic programming based speech recognizers
28. Discriminative analysis of distortion sequences in speech recognition
29. Discriminative weighting of HMM state-likelihood using the GPD method
30. Integrating syllable boundary information into speech recognition
31. Continuous speech recognition with neural networks and stationary-transitional acoustic units
32. Recognition of Mandarin syllables based on the distribution of two-dimensional cepstral coefficients
33. Finding structure in time
34. Isolated-utterance speech recognition using hidden Markov models with bounded state duration
35. Efficient search using phone probability estimates