Seminar: Speech Recognition and Natural Language Understanding

Seminar: Speech Recognition and Natural Language Understanding
Speaker: Mr. Li, Jin Cheng
Time: 10:15am - 12:15am, June 5, 1999
Place: Norcross Public Library,5570 Spalding Drive, Norcross, GA 30092
(770) 729-0931

Photos:

• LI, Jincheng in the presentation
• LI, Jincheng demonstrates voice-activated dialing with an audience, MQ.

1. Fundamental of Speech Recognition and Natural Language Understanding
   • Speech Recognition
     • Voice signal representation
     • Voice signal processing
     • Feature extraction
     • Pattern matching (Lexicon and Language Model)
   • Natural Language Understanding
     • Syntactic processing
     • Semantic Interpretation
     • Grammar
     • Context and world knowledge
2. Speech Recognition Applications
   • Command-based system
     • Replacement for IVR system
     • Directory assistance
     • Voice activated dialing
     • Stock quote
     • General information & news
     • Voice activated Web browser
   • Conversation system
     • Charles Schwab TeleBroker
     • E*Trade
     • Restaurant guide
     • Other prototype systems
   • Internet voice browser
     • VXML ( Voice eXtensible Markup Language)
3. Text To Speech
   • Synthetic Speech
   • Concatenated Speech
4. Designing Effective Speech Systems
   • When to use speech
   • Transience: What did you say?
   • Invisibility: What can I say?
   • Asymmetry
   • Speech synthesis quality
   • Speech recognition performance
   • Recognition: flexibility vs. accuracy
   • Prompting
   • Error Handling
5. Major Player
   • IBM (ViaVoice)
   • AT&T (TrueVoice, VXML)
   • Lucent (Lucent Speech Processing Solution, VXML, TTS)
   • Motorola (VOXML, VXML)
   • Philips Electronics (FreeSpeech)
   • Unisys (NLU)
   • Microsoft (Reco & NLU & TTS)
   • Nuance communications (Reco & NLU)
   • SpeechWorks (Reco & NLU)
   • Intellivoice (Voice activated dialing)
   • MIT, Lab for computer science
   • Pennsylvania State University
   • Oregon Graduate institute, Center for Spoken Language Understanding
6. Future Directions
   • Conversational systems
   • Internet voice browser