Audio-Visual Speech Recognition by Machines

In this section, we describe some existing automatic speech recognition systems based mainly on optical and acoustical information flows. We will finally present some results obtained with the automatic audio-visual speech recognizer implemented at ICP.

• Audio-visual speech perception by humans
• Automatic visual speech recognition
• Automatic audio-visual speech recognition
• Current results obtained at ICP
  • Experimental paradigm
  • Technique used
  • Preliminary results
    • Results with clear acoustic training
  • Results with degraded acoustic training
• Forecast for future works

Esprit Project 8579/MIAMI (Schomaker et al., '95)
Thu May 18 16:00:17 MET DST 1995