20-04-2011, 09:15 AM
[attachment=12432]
Speech Recognition
Introduction
• What is Speech Recognition?
- Voice Recognition?
• Where can it be used?
- Dictation
- System control/navigation
- Commercial/Industrial applications
- Hand held digital recorders
Continuous or Discrete?
• Continuous speech
- dictation
• Discrete speech
- system controls
How does SR work?
• Recognition
• Training
• Correction
• Command/Control
Recognition (1)
Recognition (2)
Acoustic Modeling
• Spoken words: “I think there are…..”
• Phonemes: ‘ ay th-in-nk-kd dh-eh-r aa-r’
• H.M.M.’s: 5 state representation
• Speech Engine
Recognition (3)
Language Modeling
• Word context
• Word frequency
• Transition possibilities
Voice Training (1)
Can be done by:
• Predetermined text segments
• Individual words
Compare new acoustic with old and combines
• More training = better recognition
Voice Training (2)
User specific Voice file
• Voice qualities
• Pronunciation
• Patterns of word use
• Preferred vocabulary
Making Corrections
• Move cursor by voice command
• Memorize edit commands
• List of possible alternatives
• Make correction manually
Command/Control
• Desktop grid
• Program or Link name/number
• URL name
• Memorized commands
Recent Improvements in SR
• Faster training ~10 min.
• Better recognition ~95%
• More compatible software
• Better system control/command
Current Software Options for PC
• Dragon Systems – Naturally Speaking
• Philips – FreeSpeech
• IBM – ViaVoice
• Lernout & Hauspie – Voice Xpress
How well do the work?
Future of SR
• SUI – Speech-based User Interface
• Improvements needed:
- Greater accuracy
- Greater system control/command
- More compatible software
Conclusion
• SR Uses
• How does it work?
• Current Software
• Problems of SR
• More SR coming soon….
Speech Recognition
Introduction
• What is Speech Recognition?
- Voice Recognition?
• Where can it be used?
- Dictation
- System control/navigation
- Commercial/Industrial applications
- Hand held digital recorders
Continuous or Discrete?
• Continuous speech
- dictation
• Discrete speech
- system controls
How does SR work?
• Recognition
• Training
• Correction
• Command/Control
Recognition (1)
Recognition (2)
Acoustic Modeling
• Spoken words: “I think there are…..”
• Phonemes: ‘ ay th-in-nk-kd dh-eh-r aa-r’
• H.M.M.’s: 5 state representation
• Speech Engine
Recognition (3)
Language Modeling
• Word context
• Word frequency
• Transition possibilities
Voice Training (1)
Can be done by:
• Predetermined text segments
• Individual words
Compare new acoustic with old and combines
• More training = better recognition
Voice Training (2)
User specific Voice file
• Voice qualities
• Pronunciation
• Patterns of word use
• Preferred vocabulary
Making Corrections
• Move cursor by voice command
• Memorize edit commands
• List of possible alternatives
• Make correction manually
Command/Control
• Desktop grid
• Program or Link name/number
• URL name
• Memorized commands
Recent Improvements in SR
• Faster training ~10 min.
• Better recognition ~95%
• More compatible software
• Better system control/command
Current Software Options for PC
• Dragon Systems – Naturally Speaking
• Philips – FreeSpeech
• IBM – ViaVoice
• Lernout & Hauspie – Voice Xpress
How well do the work?
Future of SR
• SUI – Speech-based User Interface
• Improvements needed:
- Greater accuracy
- Greater system control/command
- More compatible software
Conclusion
• SR Uses
• How does it work?
• Current Software
• Problems of SR
• More SR coming soon….