Browsing by Author "Chiang, SM"

Now showing 1 - 3 of 3

An RNN-based preclassification method for fast continuous Mandarin speech recognition
(1998-01-01)
Chen, SH
;
Liao, YF
;
Chiang, SM
;
Chang, SG
;
電信工程研究所
A novel recurrent neural network-based (RNN-based) front-end preclassification scheme for fast continuous Mandarin speech recognition is proposed in this paper, First, an RNN is employed to discriminate each input frame for the three broad classes of initial, final, and silence, A finite state machine (FSM) is then used to classify the input frame into four states including three stable states of Initial (I), Final (F), and Silence (S), and a Transient (T) state, The decision is made based on examining whether the RNN discriminates well between classes, We then restrict the search space for the three stable states in the following DP search to speed up the recognition process, Efficiency of the proposed scheme was examined by simulations in which we incorporate it with a hidden Markov model-based (HMM-based) continuous 411 Mandarin base-syllables recognizer, Experimental results showed that it can be used in conjunction with the beam search to greatly reduce the computational complexity of the HMM recognizer while keeping the recognition rate almost undegraded.
An RNN-based preclassification method for fast continuous Mandarin speech recognition
(1998-01-01)
Chen, SH
;
Liao, YF
;
Chiang, SM
;
Chang, SG
;
電信工程研究所
A novel recurrent neural network-based (RNN-based) front-end preclassification scheme for fast continuous Mandarin speech recognition is proposed in this paper, First, an RNN is employed to discriminate each input frame for the three broad classes of initial, final, and silence, A finite state machine (FSM) is then used to classify the input frame into four states including three stable states of Initial (I), Final (F), and Silence (S), and a Transient (T) state, The decision is made based on examining whether the RNN discriminates well between classes, We then restrict the search space for the three stable states in the following DP search to speed up the recognition process, Efficiency of the proposed scheme was examined by simulations in which we incorporate it with a hidden Markov model-based (HMM-based) continuous 411 Mandarin base-syllables recognizer, Experimental results showed that it can be used in conjunction with the beam search to greatly reduce the computational complexity of the HMM recognizer while keeping the recognition rate almost undegraded.
Price clustering in E-mini and floor-traded index futures
(2006-03-01)
Chung, HM
;
Chiang, SM
;
資訊管理與財務金融系註：原資管所+財金所
;
Department of Information Management and Finance
This article sets out to investigate price clustering in both the open-outcry (floor-traded) and electronically traded (E-mini) index futures markets of the DjlA, S&P 500, and NASDAQ-100 indices. The results show that although price clustering is ubiquitous in both the floor-traded and E-mini index futures markets, it nevertheless tends to be higher for open-outcry index futures, with the clustering in floor-traded NASDAQ-100 index futures demonstrating the highest level (97%) at zero digits. A significant increase was also found in price clustering in floor-traded index futures after the introduction of E-mini futures trading. The results tend to suggest that those trading mechanisms that involve higher levels of human participation, such as the open-outcry markets, may well lead to increased incidences of price clustering. (c) 2006 Wiley Periodicals, Inc.