I. Rogina and A. Waibel: Learning State--Dependant Stream Weights for Multi--Codebook HMM Speech Recognition Systems, Proc. ICASSP 94  Home/Search   Document Not in Database   Summary   Related Articles

....parsing ambiguities and dynamically adapt the vocabulary and language model of the recognizer based on the current discourse state. Figure 1: System Diagram We explore several approaches for the main processes. For example, we are experimenting with TDNN, MS TDNN [3] MLP, LVQ [4] and HMM s [5, 12] for acoustic modeling; n grams, word clustering, and automatic phrase detection for language modeling [6] statistically trained skipping parsing [7, 8] neural net parsing [9] and concept spotting parsing [10] for extracting the meaning; and statistical models as well as plan inferencing for ....

I. Rogina and A. Waibel: Learning State--Dependant Stream Weights for Multi--Codebook HMM Speech Recognition Systems, Proc. ICASSP 94

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC