Speech recognition using neural networks - Chapter 4 pps
Speech recognition using neural networks - Chapter 4 pps
... by a simple HMM-based recog- nizer. Figure 4. 3: Time Delay Neural Network. Integration Speech input Phoneme output B D G B D G 4. 3. NN-HMM Hybrids 63 and neural networks; the speech frames then ... speaker-dependent recognition, and 30.8% versus 40 .8% error for multi-speaker recognition. Training time was reduced to a reasonable level by using a 6 4- processor array of tra...
Ngày tải lên: 13/08/2014, 02:21
Speech recognition using neural networks - Chapter 6 pps
... 40 2 7 111 40 2 Substitutions 1% 28% 43 % 4% 28% 46 % Deletions 1% 8% 10% 2% 12% 14% Insertions 1% 4% 6% 0% 4% 3% Word Accuracy 97% 60% 41 % 94% 56% 37% Table 6.2: LPNN performance on continuous speech. 6. ... 8% Insertions 10% 5% 2% 4% Word Accuracy 37% 44 % 55% 60% Table 6.3: Performance of HMMs using a single gaussian mixture, vs. LPNN. perplexity System 7 111 40 2 HMM-1 5...
Ngày tải lên: 13/08/2014, 02:21
Speech recognition using neural networks - Chapter 1 pot
... volumes, and so on, x Speech Recognition using Neural Networks Joe Tebelskis May 1995 CMU-CS-9 5-1 42 School of Computer Science Carnegie Mellon University Pittsburgh, Pennsylvania 1521 3-3 890 Submitted ... that neural networks can indeed form the basis for a general pur- pose speech recognition system, and that neural networks offer some clear advantages over con...
Ngày tải lên: 13/08/2014, 02:21
Speech recognition using neural networks - Chapter 2 docx
... (200) M,A,R,K,E,T $ M, M A, A R, R K, K E, E T $ M A , M A R , A R K , R K E , K E T , E T $ MAR,KET MA,AR,KE,ET 1087 ,48 6,2502,986,38 14, 2715 generalized triphone (40 00) MARKET M 1 ,M 2 ,M 3 ; A 1 ,A 2 ,A 3 ; M = 3 843 ,2257,1056; A = 18 94, 1 247 ,3852; senone (40 00) 2. Review of Speech Recognition 22 2.3.3. ... a ij b j y t ( ) i ∑ = α j (t) t-1 t α i (t-1) . . . . a ij b j (y...
Ngày tải lên: 13/08/2014, 02:21
Speech recognition using neural networks - Chapter 3 potx
... ) 2 i ∑ =x j y i w ji i ∑ = y 1 y 2 y j x j w y y y 1 y 2 y 1 y 2 w 3 w 4 w 5 w 3 w 4 w 5 y 1 y 2 y 4 x 4 y 3 x 3 y 5 x 5 y k x k w 3 w 4 w 5 w k (a) (b) x j y i w ji –( ) 2 i ∑ =x j y i w ji i ∑ = x j y j x k y j w kj j ∑ = 3. Review of Neural Networks 46 By ... automatically optimized the architecture of MS-TDNNs, achiev- ing results that were competitive with state-of-the-a...
Ngày tải lên: 13/08/2014, 02:21
Speech recognition using neural networks - Chapter 5 doc
... no English. Janus performs speech trans- lation by integrating three modules — speech recognition, text translation, and speech gen- eration — into a single end-to-end system. Each of these modules ... was developed in conjunction with the Janus Speech- to -Speech Translation system at CMU (Waibel et al 1991, Osterholtz et al 1992, Woszczyna et al 19 94) . While a full discussion of...
Ngày tải lên: 13/08/2014, 02:21
Speech recognition using neural networks - Chapter 7 pdf
... training of a speech recognizer?” 7.5. Summary 143 7.5. Summary In this chapter we have seen that good word recognition accuracy can be achieved using neural networks that have been trained as speech ... recursive labeling. Figure 7. 14: A 3-state phoneme model outperforms a 1-state phoneme model. 80 82 84 86 88 90 92 94 96 98 100 word accuracy (%) 0 1 2 3 4 5 epochs 1-s...
Ngày tải lên: 13/08/2014, 02:21
Speech recognition using neural networks - Chapter 8 potx
... 40 2(a) 40 2(b) 111 HMM-1 55% HMM-5 96% 71% 58% 76% HMM-10 97% 75% 66% 82% LPNN 97% 60% 41 % HCNN 75% LVQ 98% 84% 74% 61% 83% TDNN 98% 78% 72% 64% MS-TDNN 98% 82% 81% 70% 85% Table 8.1: Comparative ... three conversations (41 sentences), which were used for testing; perplexity 40 2(a) used no gram- mar with the full vocabulary and again tested only the first three conversations (41 sen-...
Ngày tải lên: 13/08/2014, 02:21
Speech recognition using neural networks - Chapter 9 pptx
... 81–89, 94, 14 7-1 48 basic operation 81–82 training & testing procedures 8 2-8 4 experiments 8 4- 8 7 extensions 8 9-9 4 weaknesses 9 4- 9 9 vs. HMM 88–89, 14 7-1 48 LVQ 33, 36, 39, 54, 75, 8 8-8 9, 14 7-1 48 M maximum ... 16, 61, 80, 94, 13 8-1 43 recognition 1 4- 1 9, 5 5-5 6, 61, 8 4- 6 , 138, 14 3-1 44 , 14 8-1 49 spotting 6 9-7 1 transi...
Ngày tải lên: 13/08/2014, 02:21
speech recognition using neural networks
... performance. We will see that neural networks help to avoid this problem. 1.2. Neural Networks Connectionism, or the study of artificial neural networks, was initially inspired by neuro- biology, but it ... is active. 3.3. A Taxonomy of Neural Networks Now that we have presented the basic elements of neural networks, we will give an over- view of some different types of net...
Ngày tải lên: 28/04/2014, 10:18