1. Spoken Dialogue Systems
Julia Hirschberg
CS 4706
9/22/2018

2. Issues Error avoidance Error detection
From the system side: how likely is it the system made an error?
From the user side: what cues does the user provide to indicate an error?
Error handling: what can the system do when it thinks an error has occurred?
Evaluation: how do you know what needs fixing most?
9/22/2018

3. Avoiding misunderstandings
By imitating human performance
Timing and grounding (Clark ’03)
9/22/2018

4. Recognizing Problematic Dialogues
Hastie et al, “What’s the Trouble?” ACL 2002.
9/22/2018

5. Recognizing Problematic Utterances (Hirschberg et al ’99--)
Collect corpus from interactive voice response system
Identify speaker ‘turns’
incorrectly recognized
where speakers first aware of error
that correct misrecognitions
Identify prosodic features of turns in each category and compare to other turns
Use Machine Learning techniques to train a classifier to make these distinctions automatically
9/22/2018

6. Turn Types
TOOT: Hi. This is AT&T Amtrak Schedule System. This is TOOT. How may I help you?
User: Hello. I would like trains from Philadelphia to New York leaving on Sunday at ten thirty in the evening.
TOOT: Which city do you want to go to?
User: New York.
misrecognition
Here are examples of the 3 turn types we focus on.
correction
aware site
9/22/2018

7. Results Reduced error in predicting misrecognized turns to 8.64%
Error in predicting ‘awares’ (12%)
Error in predicting corrections (18-21%)
9/22/2018

8. Evidence from Human Performance
Users provide explicit positive and negative feedback
Corpus-based vs. laboratory experiments – do these tell us different things?
Bell & Gustafson ’00
What do we learn from this?
What functions does feedback serve?
Krahmer et al
‘go on’ and ‘go back’ signals in grounding situations (implicit/explicit verification)
Signalling whether information is grounded or not (Clark & Wilkes-Gibbs ‘86, Clark & Schaeffer ‘89): presentation/acceptance
120 dialogue for Dutch train info; one version uses explicit verification and oneimplicit; 20 users given 3 tasks; analyzed 443 verification q/a pairs
predicted that responses to correct verifications would be shorter, with unmarked word order, not repeating or correcting information but presenting new information (positive cues) -- principle of least effort
findings: where problems, subjects use more words (or say nothing), use marked word order (especially after implicit verifs), contain more disconfirmations (duh), with more repeated and corrected info
ML experiments (memory based learning) show 97% correct prediction from these features (>8 words or marked word order or corrects info -> 92%)
Krahmer et al ‘99b predicted additional prosodic cues for neg signals: high boundary tone, high pitch range, long duration of ‘nee’ and entire utterance, long pause after ‘nee’, long delay before ‘no’, from 109 negative answers to ynqs of 7 speakers; hyp
9/22/2018

9. Hypotheses supported but…
Pos: short turns, unmarked word order, confirmation, answers, no corrections or repetitions, new info
Neg: long turns, marked word order, disconfirmation, no answer, corrections, repetitions, no new info
Hypotheses supported but…
Can these cues be identified automatically?
How might they affect the design of SDS?
9/22/2018

10. Error Handling Strategies
Goldberg et al ’03: how should systems best inform the user that they don’t understand?
System rephrasing vs. repetitions vs. statement of not understanding
Apologies
What behaviors might these produce?
Hyperarticulation
User frustration
User repetition or *rephrasing
9/22/2018

11. What lessons do we learn? What produces least frustration?
Best recognized input?
9/22/2018

12. Evaluating Dialogue Systems
PARADISE framework (Walker et al ’00)
“Performance” of a dialogue system is affected both by what gets accomplished by the user and the dialogue agent and how it gets accomplished
Maximize
Task Success
Minimize
Costs
Efficiency
Measures
Qualitative
Measures
9/22/2018

13. Task Success Task goals seen as Attribute-Value Matrix
ELVIS retrieval task (Walker et al ‘97)
“Find the time and place of your meeting with Kim.”
Attribute Value
Selection Criterion Kim or Meeting
Time 10:30 a.m.
Place 2D516
Task success defined by match between AVM values at end of with “true” values for AVM
9/22/2018

14. Metrics
Efficiency of the Interaction:User Turns, System Turns, Elapsed Time
Quality of the Interaction: ASR rejections, Time Out Prompts, Help Requests, Barge-Ins, Mean Recognition Score (concept accuracy), Cancellation Requests
User Satisfaction
Task Success: perceived completion, information extracted
9/22/2018

15. Experimental Procedures
Subjects given specified tasks
Spoken dialogues recorded
Cost factors, states, dialog acts automatically logged; ASR accuracy,barge-in hand-labeled
Users specify task solution via web page
Users complete User Satisfaction surveys
Use multiple linear regression to model User Satisfaction as a function of Task Success and Costs; test for significant predictive factors
9/22/2018

16. User Satisfaction: Sum of Many Measures
Was Annie easy to understand in this conversation? (TTS Performance)
In this conversation, did Annie understand what you said? (ASR Performance)
In this conversation, was it easy to find the message you wanted? (Task Ease)
Was the pace of interaction with Annie appropriate in this conversation? (Interaction Pace)
In this conversation, did you know what you could say at each point of the dialog?
(User Expertise)
How often was Annie sluggish and slow to reply to you in this conversation? (System Response)
Did Annie work the way you expected her to in this conversation? (Expected Behavior)
From your current experience with using Annie to get your , do you think you'd use Annie regularly to access your mail when you are away from your desk? (Future Use)
9/22/2018

17. Performance Functions from Three Systems
ELVIS User Sat.= .21* COMP * MRS * ET
TOOT User Sat.= .35* COMP + .45* MRS - .14*ET
ANNIE User Sat.= .33*COMP + .25* MRS +.33* Help
COMP: User perception of task completion (task success)
MRS: Mean recognition accuracy (cost)
ET: Elapsed time (cost)
Help: Help requests (cost)
9/22/2018

18. Performance Model
Perceived task completion and mean recognition score are consistently significant predictors of User Satisfaction
Performance model useful for system development
Making predictions about system modifications
Distinguishing ‘good’ dialogues from ‘bad’ dialogues
But can we also tell on-line when a dialogue is ‘going wrong’
9/22/2018

19. Next Week
Speech summarization and data mining
9/22/2018