Natural Language Processing
description
Transcript of Natural Language Processing
![Page 1: Natural Language Processing](https://reader036.fdocuments.in/reader036/viewer/2022081519/56813c88550346895da62a9f/html5/thumbnails/1.jpg)
1
Natural Language Processing
Lecture Notes 14Chapter 19
![Page 2: Natural Language Processing](https://reader036.fdocuments.in/reader036/viewer/2022081519/56813c88550346895da62a9f/html5/thumbnails/2.jpg)
2
Today
• Spoken dialogue systems
![Page 3: Natural Language Processing](https://reader036.fdocuments.in/reader036/viewer/2022081519/56813c88550346895da62a9f/html5/thumbnails/3.jpg)
3
Talking to Computers
• Spoken dialogue systems make it possible to accomplish real tasks without talking to a real person
• Keys to success– goal-directed interactions in a limited
domain– Priming users to adopt a vocabulary
you can recognize– Segmenting the task into manageable
stages– Judicious use of system vs. mixed
initiative
![Page 4: Natural Language Processing](https://reader036.fdocuments.in/reader036/viewer/2022081519/56813c88550346895da62a9f/html5/thumbnails/4.jpg)
4
Dialogue vs. Monologue• Monologue and dialogue both involve
interpreting– Information status (given and new info)– Coherence issues– Reference resolution– Speech acts, implicature, intentionality
• Dialogue involves managing– Turn-taking– Grounding – Detecting and repairing misunderstandings– Initiative and confirmation strategies
![Page 5: Natural Language Processing](https://reader036.fdocuments.in/reader036/viewer/2022081519/56813c88550346895da62a9f/html5/thumbnails/5.jpg)
5
Example
• Here’s an (unfair) example from the ATT Toot system.
![Page 6: Natural Language Processing](https://reader036.fdocuments.in/reader036/viewer/2022081519/56813c88550346895da62a9f/html5/thumbnails/6.jpg)
6
Segmenting Speech into Utterances
• What is an `utterance’?– Single syntactic sentence may span
several turnsA: We've got you on USAir flight 99B: YepA: leaving on December 1.
– Multiple syntactic sentences may occur in single turn
A: We've got you on USAir flight 99 leaving on December. Do you need a rental car?
– Utterance segmentation: cue words, n-gram word or POS sequences, prosody (pitch, accent, phrase-final lengthening, pause duration)
![Page 7: Natural Language Processing](https://reader036.fdocuments.in/reader036/viewer/2022081519/56813c88550346895da62a9f/html5/thumbnails/7.jpg)
7
Turns and Utterances
• Dialogue is characterized by turn-taking: – Who should talk next– When they should talk
• Turns in recorded speech:– Little speaker overlap (around 5% in
English)– But little silence between turns either
![Page 8: Natural Language Processing](https://reader036.fdocuments.in/reader036/viewer/2022081519/56813c88550346895da62a9f/html5/thumbnails/8.jpg)
8
Turn Taking
• How do we know when a speaker is – Giving up or taking a turn? – Holding the floor? – Interruptable?
• How do I know when– Its my turn obligatorily– Optionally?
![Page 9: Natural Language Processing](https://reader036.fdocuments.in/reader036/viewer/2022081519/56813c88550346895da62a9f/html5/thumbnails/9.jpg)
9
Simple Turn-Taking Rules• At each transition-relevance place
(TRP) of each turn:– If current speaker has selected A as next
speaker, then A must speak next– If current speaker does not select next
speaker, any other speaker may take next turn
– If no one else takes next turn, the current speaker may take next turn
• TRPs are where the structure of the language allows speaker shifts to occur
![Page 10: Natural Language Processing](https://reader036.fdocuments.in/reader036/viewer/2022081519/56813c88550346895da62a9f/html5/thumbnails/10.jpg)
10
Turn Taking Scripts• Adjacency pairs set up next speaker
expectations– GREETING/GREETING– QUESTION/ANSWER– COMPLIMENT/DOWNPLAYER– REQUEST/GRANT
• Significant silence is dispreferredA: Is there something bothering you or not?
(1.0s)A: Yes or no? (1.5s)A: Eh?B: No.
![Page 11: Natural Language Processing](https://reader036.fdocuments.in/reader036/viewer/2022081519/56813c88550346895da62a9f/html5/thumbnails/11.jpg)
11
Overall System Strategies
• System Initiative (Control freak)S: Please give me your arrival city name.U: Baltimore.S: Please give me your departure city nameU: BostonS:…
• Rigid, unnatural, difficulty with chatty users
![Page 12: Natural Language Processing](https://reader036.fdocuments.in/reader036/viewer/2022081519/56813c88550346895da62a9f/html5/thumbnails/12.jpg)
12
Overall System Strategies
• Mixed initiativeS: How may I help you?U: I want to go to Boston.S: What day do you want to go to Boston?
• User InitiativeS: How may I help you?U: I want to go from Boston to Baltimore on
November 8.
![Page 13: Natural Language Processing](https://reader036.fdocuments.in/reader036/viewer/2022081519/56813c88550346895da62a9f/html5/thumbnails/13.jpg)
13
Grounding
• Participants are trying to come to a meeting of minds, they’re trying to establish common ground (or a set of mutual beliefs)
![Page 14: Natural Language Processing](https://reader036.fdocuments.in/reader036/viewer/2022081519/56813c88550346895da62a9f/html5/thumbnails/14.jpg)
14
Grounding
• Hearers must ground a speaker’s utterances by making it clear whether or not understanding has occurred
• Various ways to do this…
S: I can upgrade you to an SUV at that rate.
User: ????
![Page 15: Natural Language Processing](https://reader036.fdocuments.in/reader036/viewer/2022081519/56813c88550346895da62a9f/html5/thumbnails/15.jpg)
15
Grounding (Clark and Schaefer 1989)
S: I can upgrade you to an SUV at that rate.– Continued attention/permission to proceed
(User gazes appreciatively at S)– Relevant next contribution
U: Do you have an Explorer available?– Acknowledgement/backchannel
U: Ok/uh-huh/Great!– Display/repetition
U: An SUV? U: You can upgrade me to an SUV at the same rate?
– Request for repairU: Huh?
![Page 16: Natural Language Processing](https://reader036.fdocuments.in/reader036/viewer/2022081519/56813c88550346895da62a9f/html5/thumbnails/16.jpg)
16
Evaluation
• Performance of a dialogue system is affected both by what gets accomplished by the user and the dialogue agent and how it gets accomplished
MaximizeMaximizeTask SuccessTask Success
Minimize Minimize CostsCosts
EfficiencyEfficiencyMeasuresMeasures
QualitativeQualitativeMeasuresMeasures
![Page 17: Natural Language Processing](https://reader036.fdocuments.in/reader036/viewer/2022081519/56813c88550346895da62a9f/html5/thumbnails/17.jpg)
17
Metrics
• Efficiency of the Interaction:User Turns, System Turns, Elapsed Time
• Quality of the Interaction: ASR rejections, Time Out Prompts, Help Requests, Barge-Ins, Cancellation Requests, …
• User Satisfaction• Task Success: perceived completion,
information extracted, information learned (in tutorial setting)
![Page 18: Natural Language Processing](https://reader036.fdocuments.in/reader036/viewer/2022081519/56813c88550346895da62a9f/html5/thumbnails/18.jpg)
18
User Satisfaction Metrics• TTS Performance
– Was system easy to understand in this conversation?• ASR Performance
– In this conversation, did system understand what you said?• Task Ease
– In this conversation, was it easy to do what you wanted?• Interaction Pace
– Was the pace of interaction appropriate in this conversation?
• User Orientation– In this conversation, did you know what you could say at
each point of the dialog? • System Response
– How often was the system sluggish and slow to reply to you in this conversation?
• Expected Behavior– Did system work the way you expected it to in this
conversation?
![Page 19: Natural Language Processing](https://reader036.fdocuments.in/reader036/viewer/2022081519/56813c88550346895da62a9f/html5/thumbnails/19.jpg)
19
Misrecognition Repair
• Recognizing when the conversation has gone astray and recovering…
– Mainly by analyzing the user’s utterances for signals that things are going astray
![Page 20: Natural Language Processing](https://reader036.fdocuments.in/reader036/viewer/2022081519/56813c88550346895da62a9f/html5/thumbnails/20.jpg)
20
Wrapup
• Spoken dialogue systems present new problems -- but also new possibilities– Recognizing speech introduces a new
source of errors– Additional information provided in the
speech stream offers new information about users’ intended meanings, emotional state (grounding of information, speech acts, reaction to system errors)