On supervised learning from sequential data with applications for speech recognition 情報系列からの教師付き学習とその音声認識への応用

Access this Article

Title: On supervised learning from sequential data with applications for speech recognition

博士論文

On supervised learning from sequential data with applications for speech recognition / p1 (0003.jp2)
Contents / p5 (0005.jp2)
List of Figures / p9 (0007.jp2)
List of Tables / p11 (0008.jp2)
List of Symbols / p14 (0009.jp2)
1 Introduction / p1 (0010.jp2)
1.1 MOTIVATION AND BACKGROUND / p1 (0010.jp2)
1.2 THESIS STRUCTURE / p5 (0012.jp2)
2 Supervised learning from sequential data / p7 (0013.jp2)
2.1 DEFINITION OF THE PROBLEM / p7 (0013.jp2)
2.2 Decomposition into a generative and a prior model part / p10 (0015.jp2)
2.3 Direct decomposition / p13 (0016.jp2)
2.4 HIDDEN MARKOV MODELS / p14 (0017.jp2)
2.5 SUMMARY / p23 (0021.jp2)
3 Neural networks for supervised learning from sequences / p25 (0022.jp2)
3.1 BASICS OF NEURAL NETWORKS / p26 (0023.jp2)
3.2 BIDIRECTIONAL RECURRENT NEURAL NETWORKS / p36 (0028.jp2)
3.3 MIXTURE DENSITY RECURRENT NEURAL NETWORKS / p53 (0036.jp2)
3.4 SUMMARY / p65 (0042.jp2)
4 Memory-efficient LVCSR search using a one-pass stack decoder / p69 (0044.jp2)
4.1 INTRODUCTION / p70 (0045.jp2)
4.2 A MEMORY-EFFICIENT ONE-PASS STACK DECODER / p80 (0050.jp2)
4.3 EXPERIMENTS / p97 (0058.jp2)
4.4 CONCLUSIONS / p105 (0062.jp2)
4.5 ACKNOWLEDGMENTS / p105 (0062.jp2)
5 Conclusions / p107 (0063.jp2)
5.1 SUMMARY / p107 (0063.jp2)
5.2 CONTRIBUTIONS FROM THIS THESIS / p109 (0064.jp2)
5.3 SUGGESTIONS FOR FUTURE WORK / p110 (0065.jp2)
A Improved initialization of Multi-Layer-Perceptrons / p113 (0066.jp2)
B List of Publications / p115 (0067.jp2)

7access