J.H. Kim and S.B. Lee
Multilayer recurrent neural prediction model, neural network, hidden Markov model, VQ, speech recognition, DSP
In this paper, the recurrent neural networks (RNNs) are applied to compensate for hidden Markov model (HMM) recognition algorithm, which is commonly used as a main recognizer. Among these RNNs, the multilayer recurrent neural prediction model (MRNPM), which allows operating in real-time, is used to implement learning and recognition, and HMM and MRNPM are used to design a hybrid- type main recognizer. After testing the designed speech recognition algorithm with Korean number pronunciations (13 words), which are hardly distinct, for its speech-independent recognition ratio, about a 5% improvement is obtained comparing results with existing HMM recognizers. Based on this result, only optimal (recognition) codes were extracted in the actual DSP (TMS320C6711) environment and the embedded speech recognition system was implemented. Similarly, the implementation result of the embedded system showed an improved recognition system implementation than existing solid HMM recognition systems.
Important Links:
Go Back
