Error Correction Techniques for Handwriting, Speech, and other ambiguous or error prone systems

Jennifer Mankoff & Gregory D. Abowd GVU Center & College of Computing
Georgia Institute of Technology, Atlanta, GA, USA
+1 404 894 7512
jmankoff@cc.gatech.edu, abowd@cc.gatech.edu
http://www.cc.gatech.edu/fce/pendragon

Abstract:

Interfaces which support natural inputs such as handwriting and speech are becoming more prevalent. However, these recognition-based interface techniques are error prone. Despite research efforts to improve recognition rates, a certain amount of error will never be removed. Suitable research efforts should attend to the problem of correction techniques for these error prone techniques. Humans have developed countless ways to correct errors in understanding or clarify ambiguous statements. It is time for interface designers to focus on ways for computers to do the same. We present a survey of the design, implementation, and study of interfaces for correcting error prone input technologies. Previous work by others and our own research into flexible pen-based note-taking environments grounds our research into interface techniques for handling errors in recognition systems.

handwriting and speech recognition, interface design, error handling

• 1 INTRODUCTION
  • 1.1 Motivating the Problem
  • 1.2 Defining The Area
• 2 ERROR DISCOVERY
  • 2.1 User input to help the system find errors
  • 2.2 System output to help the user find errors
  • 2.3 Thresholding
  • 2.4 Rules
  • 2.5 Historical Statistics
• 3 ERROR CORRECTION TECHNIQUES
  • 3.1 Choosing a Default
  • 3.2 Encouraging Less Ambiguous Input
  • 3.3 Mimicking Natural Human Correction Strategies
• 4 VALIDATION OF TECHNIQUES
• 5 TOOLKIT LEVEL SUPPORT FOR ERROR HANDLING
• 6 CONCLUSION
• References