Introduction to computational linguistics
Jay Munson(special thanks to Misty
Azara)
May 30, 2003
Today’s Goals I. Introduction to computational
linguistics (CL) through the discussion of 7 CL core areas.
II. Identify Common CL applications III. Identify the importance of
theoretical linguistics in CL
What is computational linguistics? Essentially, CL is any task, model,
algorithm, etc. that attempts to place any type of language processing (syntax, phonology, morphology, etc.) in a computational setting
What is computational linguistics (CL)? CL is interdisciplinary
Linguistics Computer Science Mathematics Electrical Engineering Speech and Hearing Science
Seven Core Areas of CL 1. Machine Translation 2. Speech Recognition 3. Text-to-Speech 4. Natural Language Generation 5. Human-Computer Dialogs 6. Information Retrieval 7. Computational Modeling
1.0 Machine Translation (MT)
Using computers to automate some or all of translating from
one language to another
1.1 MT (cont.) Three general models or tasks:
Tasks for which a rough translation is adequate
Tasks where a human post-editor can be used to improve the output
Tasks limited to a small sublanguage
1.2 MT (cont.) Linguistic knowledge is extremely
useful in this area of CL MT benefits from knowledge of
language typology and language-specific linguistic information
Programs are typically “trained” using pre-translated documents/texts.
1.3 MT Example KANT Knowledge-based Machine Translation The KANT project, Knowledge-based, Accurate Translation
for technical documentation, was founded in 1989 for the research and development of large-scale, practical translation systems for technical documentation. KANT uses a controlled vocabulary and grammar for each source language, and explicit yet focused semantic models for each technical domain to achieve very high accuracy in translation. Designed for multilingual document production, KANT has been applied to the domains of electric power utility management and heavy equipment technical documentation.
http://www.lti.cs.cmu.edu/Research/cmt-projects.html
2.0 Speech Recognition (SR)
Taking spoken language as input and outputting the
corresponding text
2.1 SR - Architecture SR takes the source speech and
produces “guesses” as to which words could correspond to the source via some type of acoustic model
The word with the highest probability is selected as the optimal candidate
Contexts are “contained” to improve accuracy
2.2 Why use SR? Allow for hands-free human-
computer interaction Assists in automated telephony
3.0 Text-to-Speech (TTS)
Taking text as input and outputting the corresponding
spoken language
3.1 Three types of TTS 1. Articulatory- models the physiological
characteristics of the vocal tract 2. Concatenative- uses pre-recorded
segments to construct the utterance(s) ScanSoft: Jennifer and Susana
http://www.scansoft.com/realspeak/demo/ Speechify: British Female
http://www.speechworks.com/demos/speechify.cfm
3.2 Three types of TTS (cont.) 3. Parametric/Formant- models the
formant transitions of speech ETI-Eloquence: Reed
http://www.speechworks.com/demos/eti.cfm
3.3 Why is TTS so difficult? Spelling
through, rough, though, thought Homonyms
PERmit (n) vs. perMIT (v) Prosody (dependent on context)
Pitch, duration of segments, phrasing of segments, intonational tune, emotion
“I am so angry at you. I have never been more enraged in my life!!”
3.4 Why use TTS? Allows for text to be read automatically Extremely useful for the visually and
hearing impaired.
For a review of the history of TTS until 1987 with sound files, goto:
http://www.ece.ogi.edu/~macon/ECE580/klatt/
4.0 Natural Language Generation (NLG)
Constructing linguistic outputs from non-linguistic inputs; the NLG goal is to produce natural language
from internal data/structure.
4.1 Natural language generation (cont) Maps meaning to text Nature of the input varies greatly
from one application to another (i.e documenting structure of a computer program)
The job of the NLG system is to extract the necessary information to drive the generation process
4.2 NLG systems have to make choices: Content selection- the system must
choose the appropriate content for input, basing its decision on a pre-specified communicative goal
Lexical selection- the system must choose the lexical item most appropriate for expressing a concept
4.3 NLG (cont) Sentence Structure
Aggregation- the system must apportion the content into phrase, clause, and sentence-sized chunks
Referential expression- the system must determine how to refer to the objects under discussion (not a trivial task).
4.4 NLG - Structures Discourse structure- many NLG
systems have to deal with multi-sentence discourses, which must have a coherent structure
4.5 Sample NLG outputTo save a file
1. Choose save from the file menu2. Choose the appropriate folder3. Type the file name4. Click the save button
The system will save the document.…
5.0 Human-Computer Dialogs
Uses a mix of SR, TTS, and pre-recorded prompts to
achieve some goal
5.1 Human-Computer Dialogs Uses speech recognition, or a
combination of SR and touch tone as input to the system
The system processes the spoken information and outputs appropriate TTS or pre-recorded prompts
5.2 Human-Computer Dialogs Dialog systems have specific tasks,
which limit the domain of conversation
This makes the SR problem much easier, as the potential responses become very constrained
5.3 Sample dialog system for banking…Sys: would you like information for
checking or savings? User: Checking, please.Sys: Your current balance is $2,568.92.
Would you like another transaction?User: Yes, has check #2431 cleared?…
5.4 Linguistic knowledge in dialog systems Discourse structure- ensuring natural
flowing discourse interaction Building appropriate
vocabularies/lexicons for the tasks Ensuring prosodic consistencies (i.e.
questions sound like questions and spliced prompts sound continuous)
5.5 Why use human-computer systems? Automate simple tasks- no need
for a teller to be on the other end of the line!
Allow access to system information from anywhere, via the telephone
6.0 Information Retrieval
Storage, analysis, and retrieval of text documents
6.1 Information Retrieval (IR) Most current IR systems are based
on some interpretation of “compositional semantics” (e.g. the meaning of the whole is based the meaning of its parts and their combination).
IR is the core of web-based searching, i.e. Google, Altavista, etc.
6.2 IR - Architecture User inputs a word or string of
words System processes the words and
retrieves documents corresponding to the request
6.3 “Bag of Words” The dominant approach to IR
systems is to ignore syntactic information and process the meaning of individual words only
Thus, “I see what I eat” and “I eat what I see” would mean exactly the same thing to the system!
6.4 Linguistic Knowledge in IR Semantics
Compositional Lexical
Syntax (depending on the model used)
7.0 Computational Modeling
Computational approaches to problem solving, modeling,
and development of theories
7.1 How can we use computational modeling? Develop working models of
language evolution Model speech perception,
production, and processing Almost any theoretical model can
have a computational counterpart
7.2 Why Use Computational Modeling? Forces explicitness – no black boxes
or behind the scenes “magic” Allows us to test our formal theories
given a large amount of data Allows for enhancements in
technology and benefits to society through the implementaions of models.
Conclusions CL applications utilize linguistic
knowledge from all of the major subfields of theoretical linguistics (e.g. theory is necessary!)
Computational modeling can aid/test linguists’ theories of language processing and structure
Conclusions - Review of 7 core areas in CL 1. Machine Translation 2. Speech Recognition 3. Text-to-Speech 4. Natural Language Generation 5. Human-Computer Dialogs 6. Information Retrieval 7. Computational Modeling
Conclusions – Review of Today’s Goals I. Introduction to computational
linguistics (CL) through the discussion of 7 CL core areas.
II. Identify Common CL applications III. Identify the importance of
theoretical linguistics in CL
El fin.