Dialog Design

1
Dialog Design

• 1. WIMP
• 2. Command language
• 3. Direct manipulation
• 4. Speech

2
Direct Manipulation

• 1) Continuous visibility of the objects and
  actions of interest
• 2) Rapid, reversible, incremental actions
• 3) Replacement of command language syntax by
  direct manipulation of object of interest

3
Direct Manipulation

• Examples
• WYSIWYG editors and word processors
• VISICALC - 1st electronic spreadsheet
• CAD
• Desktop metaphor
• Video games

4
DM Essence

• Representation of reality that can be manipulated
• The user is able to apply intellect directly to
  the task
• The tool itself seems to disappear

5
Direct Manipulation

• Advantages

• Disadvantages

6
DM Problems

• Waste of screen space (info not all that dense)
• Need to learn meaning of components of visual
  representation
• Visual representation may be misleading
• Mouse ops may be slower than typing
• Not self-explanatory (no prompts)

7
DM Problems

• Not good at
• Repetition
• History keeping (harder)
• Certain tasks (Change all italics to bold)
• Macros harder

8
DM Advantages

• Easier to learn remember
• Direct WYSIWYG
• Flexible, easily reversible actions
• Provides context instant visual feedback
• Exploits human use of visual spatial cues
• Limits types of errors

9
What is DM?

• Word? Emacs?
• In end, must characterize direct manipulation by
  feeling of directness and illusion of
  manipulating objects at hand

10
Speech Input

• Speech synthesis
• Speaker recognition
• Speech recognition
• Natural language understanding

11
English Speech

• Made up of 40 phonemes, 24 consonants and 16
  vowels

12
Speech Synthesis

• Often hear robotic voice
• Store tones, then put them together
• -gt The transition is the difficult thing to do

13
Speaker Recognition

• Tell which person it is (voice print)
• Could be important for monitoring meetings

14
Speech Recognition

• Primarily identifying words
• Improving all the time
• Commercial systems
• IBM ViaVoice, Dragon Dictate, ...

15
Recognition Dimensions

• Speaker dependent/independent
• Parametric patters are sensitive to speaker
• With training (dept) can get better
• Vocabulary
• Some are getting 50,000 words
• Isolated word vs. continuous speech
• Continuous where words stop begin
• Typically a pattern match, no context used

16
Recognition Systems

• Typical system has 5 components
• Speech capture device - Has analog -gt digital
  converter
• Digital Signal Processor - Gets word boundaries,
  scales, filters, cuts out extra stuff
• Preprocessed signal storage - Processed speech
  buffered for recognition algorithm
• Reference speech patterns - Stored templates or
  generative speech models for comparisons
• Pattern matching algorithm - Goodness of fit from
  templates/model to users speech

17
Errors

• Systems make four types of errors
• Substitution - one for another
• Rejection - detected, but not recognized
• Insertion - added
• Deletion - not detected
• Which is more common, dangerous?

18
Natural Language Understanding

• Putting meaning to the words
• Input might be speech or could be typed in
• Holy grail of Artificial Intelligence problems

19
NL Factors/Terms

• Syntactic
• Grammar or structure
• Prosodic
• Inflection, stress, pitch, timing
• Pragmatic
• Situated context of utterance, location, time
• Semantic
• Meaning of words

20
SR/NLU Assessment

• Advantages

• Disadvantages

21
SR/NLU Advantages

• Easy to learn and remember
• Powerful
• Fast, efficient (not always)
• Little screen real estate

22
SR/NLU Disadvantages

• Doesnt work good enough yet
• Assumes knowledge of problem domain
• Not prompted, like menus
• Requires typing skill (if keyboard)
• Enhancements are invisible
• Expensive to implement