20 Aug l at CUHK A Feature Weighting Method for Robust speech Recognition Speech Activities in CST Thomas Fang Zheng Center of Speech Technology State Key Lab of Intelligent Technology and systems Department of Computer Science Technology Tsinghua University fzheng@sp.cs.tsinghua.edu.cn,http:/sp.cs.tsinghuaeducn/fzheng
A Feature Weighting Method for Robust Speech Recognition Thomas Fang Zheng Center of Speech Technology State Key Lab of Intelligent Technology and Systems Department of Computer Science & Technology Tsinghua University fzheng@sp.cs.tsinghua.edu.cn, http://sp.cs.tsinghua.edu.cn/~fzheng/ -- Speech Activities in CST 20 Aug 01 at CUHK
Center of Speech Technology a Founded in 1979, named as Speech laboratory a Joined the State Key laboratory of intelligent Technology and Systems in 1999. renamed as Center of speech Technolog ahttp:/sp.cs.tsinghuaeducn/ Center of speech Technology, Tsinghua University Slide 2
Center of Speech Technology, Tsinghua University Slide 2 ❑ Founded in 1979, named as Speech Laboratory ❑ Joined the State Key Laboratory of Intelligent Technology and Systems in 1999, renamed as Center of Speech Technology ❑ http://sp.cs.tsinghua.edu.cn/ Center of Speech Technology
Members of cst in 2001 5 ■ Post doctors 口 Doc tora1 Students 口 Master students Center of speech Technology, Tsinghua University Slide 3
Center of Speech Technology, Tsinghua University Slide 3 5 1 13 7 Faculty Post Doctors Doctoral Students Master Students Members of CST in 2001
Founding resources a State fundamental research plan: NSF, 863, 973, 985 o Collaboration with industries 令 Microsoft 令IBM 冷 Intel 今 Lucent technologies ☆NOka 令 Wenwen SoundTek ☆ keysun Center of speech Technology, Tsinghua University Slide 4
Center of Speech Technology, Tsinghua University Slide 4 ❑ State fundamental research plan: NSF, 863, 973, 985 ❑ Collaboration with industries: ❖ Microsoft ❖ IBM ❖ Intel ❖ Lucent Technologies ❖ Nokia ❖ Weniwen ❖ SoundTek ❖ Keysun ❖ ... Founding Resources
Speech Research Activities a Acoustic Modeling Natural/Spoken Speech s Feature Extraction and Selection Understanding(NLU/SLU 今 Acoustic Modeling e NLU -GLR Based Parsing Accurate fast AM Search , SLU-Kw based robust parsing 令 Robustness 令 Dialogue Manager Speech Enhancement 日 Applications Fractals ☆ Command and control Speaker Adaptation Speaker Normalization Keyword spotting Chinese Pronunciation Modeling 令 Language Learning a Language modeling Input method editor Chinese dictation machine Characteristics of Chinese 令 Spoken dialogues ,s Language Modeling and Search Speaker identification and , LM Adaptation New Word verification Induction 口 Resources Center of speech Technology, Tsinghua University Slide 5
Center of Speech Technology, Tsinghua University Slide 5 Speech Research Activities ❑ Acoustic Modeling ❖ Feature Extraction and Selection ❖ Acoustic Modeling ❖ Accurate & fast AM Search ❖ Robustness ➢ Speech Enhancement ➢ Fractals ➢ Speaker Adaptation ➢ Speaker Normalization ➢ Chinese Pronunciation Modeling ❑ Language Modeling ❖ Characteristics of Chinese ❖ Language Modeling and Search ❖ LM Adaptation & New Word Induction ❑ Natural/Spoken Speech Understanding (NLU/SLU) ❖ NLU - GLR Based Parsing ❖ SLU - KW based robust parsing ❖ Dialogue Manager ❑ Applications ❖ Command and control ❖ Keyword spotting ❖ Language Learning ❖ Input method editor ❖ Chinese dictation machine ❖ Spoken dialogues ❖ Speaker identification and verification ❑ Resources
Feature extraction and selection a Trying to extract discriminative features Trying to select robust feature components from the existing features Center of speech Technology, Tsinghua University Slide 6
Center of Speech Technology, Tsinghua University Slide 6 ❑ Trying to extract discriminative features ❑ Trying to select robust feature components from the existing features Feature Extraction and Selection
a Introduction human does not always use an identical feature to recognize objects s It is often that the feature components being used vary with the objects to be recognized This is feature selection after the feature extraction feature selection can be regarded as a special case of feature weighting Before going to the topic let's see a problem at first Center of speech Technology, Tsinghua University Slide 7
Center of Speech Technology, Tsinghua University Slide 7 ❑ Introduction ❖ Human does not always use an identical feature to recognize objects. ❖ It is often that the feature components being used vary with the objects to be recognized. ❖ This is feature selection after the feature extraction. ❖ Feature selection can be regarded as a special case of feature weighting. ❖ Before going to the topic, let’s see a problem at first
Room 1 Room 2 o0O0。O m八m B X 口 A Problem 冷 Conditions Two opaque and completely separated rooms are very close to each other Room 1 contains 3 switches and room 2 contains 3 lights Each switch is corresponding to one and only one light You can switch on/off any switch for any times But you can only enter each room for only once , Goal: finding which switch is corresponding to which light Center of speech Technology, Tsinghua University Slide 8
Center of Speech Technology, Tsinghua University Slide 8 ❑ A Problem ❖ Conditions: ➢ Two opaque and completely separated rooms are very close to each other. ➢ Room 1 contains 3 switches and Room 2 contains 3 lights. ➢ Each switch is corresponding to one and only one light. ➢ You can switch on/off any switch for any times. ➢ But you can only enter each room for only once. ❖ Goal: finding which switch is corresponding to which light. Room 1 Room 2 A B C X Y Z
Room 1 Room 2 “ O0 OOOO m八m B X Y a The answer(Step 1) Unseen 令 Actions Turn on switch a Wait fora couple of minutes Turn off Switch A Center of speech Technology, Tsinghua University Slide g
Center of Speech Technology, Tsinghua University Slide 9 Unseen A B C X Y Z Room 1 Room 2 ❑ The Answer (Step 1) ❖ Actions: ➢ Turn on Switch A. ➢ Wait for a couple of minutes ... ➢ Turn off Switch A
Room 1 Room 2 T/( B X Y 口 The answer(Step2) Unseen 令 Actions: Turn on Switch B Immediately go to Room 2 Center of speech Technology, Tsinghua University Slide 10
Center of Speech Technology, Tsinghua University Slide 10 Unseen A B C X Y Z Room 1 Room 2 ❑ The Answer (Step 2) ❖ Actions: ➢ Turn on Switch B. ➢ Immediately go to Room 2