Lecture 6-1

Introduction to Language Technologies:
Challenges and Applications

Sameh Alansary
[email protected]
Prof. of Computational Linguistics

Head of Phonetics and Linguistics Dept.
Faculty of Arts, Alexandria University
Lecture 6
Applications of Computational linguistics
Primary Applications
• A primary applications has always been the development of specific practical systems which involve
natural language. Three important applications on development of computational linguistics are:
1) Machine Translation
2) Information Retrieval (IR)
3) Man-Machine Interface/ Human Computer Interaction (HCI)

Secondary Applications: Text Processing
❖Computer- Aided Language Learning (CALL)/ Computer- Aided Language
Teaching (CALT)
❖Corpus creation and documentation E.g., CIIL- LDC- IL Corpora
❖Word Sense Disambiguation (WSD) E.g., Machine Learning Algorithms
❖Natural language Question and Answering (QA) System
❖Optical Character Recognition (OCR)
❖Sentiment Analysis/ Emotional Identification
❖Information Extraction (IE) E.g., Google Search engine

Secondary Applications: Text Processing
❖Automatic Text Summarization
❖Spelling & Grammar checking
❖Plagiarism Detection
❖Automatic Morphological analysis and generation
❖Named Entity Recognition (NER)
❖Parts- of- Speech Tagging / Annotation
❖Text-to-Speech
Secondary Applications: Speech Processing
❖Speech Recognition E.g.. Speech- To- Text (STT)
❖Speech Synthesis E.g., Speech- To- Speech (STS)
❖Speech enabled search engine. E.g., Google voice enabled search engine
❖Speaker identification
❖Natural Language Dialogue Interfaces to Databases
❖Spoken dialogue system
❖Speech corpus creation and development
❖Voice dialing system
❖Automatic phonetic dictionary

Machine Translation
• The purpose of the machine translation system is taking text written or spoken in one
language and writing or speaking it in another
• Researchers not attempting to build full-fledged machine translation system
• They are attempting to build some projects that aiming towards machine-assisted
translation
• The computer is viewed as a tool to aid the translator
• The computer makes suggestions but the human make final decisions
Source Machine Target
language translation language
system
Spoken Speech Written Machine Written Speech Spoken

source recognition source translation target synthesis target
language system language system language system language
Speech Recognition
• Definition: Speech recognition turns acoustic input into strings of phonemes
and then finds the best matching word in a database.
• Can be built for open domain use, theoretically recognizing all possible strings of words
• e.g. dictation systems
• Can also be built for a particular domain, recognizing small, finite sets of utterances
• e.g. automated call-centers.

Silent Speech Recognition
Natural Language Generation (NLG)
Constructing linguistic outputs from non-linguistic inputs; the NLG

goal is to produce natural language from internal data/structure.
“I'm sorry, Dave, I'm afraid I can't do that”

pragmatics:
politeness
indirect speech
morphology:
contractions
discourse:
reference (“that”)
Speech Synthesis
• Has the opposite goal from a speech recognition system
• One of the most important uses is reading to visually impaired
Speech
Written Spoken
synthesis
language language
system
Information Access and Retrieval
• The internet and the world wide web have created a growing body of text and images
that can now be searched by any one
• There are dozens of freely available search systems or engines for roaming through
titles, articles , caption and other sections of texts
• If linguistically sophisticated program had been used to retrieve these titles it is likely
that they would have been divided according to the semantic subject field
Indexing and Concordances
• Indexing means finding, identifying, and usually counting all occurrences of a certain
word in large texts
• Computers are best at doing : locating a word , recording the location by line or
sentence number , and counting how many times it appears
• A concordance tells which words occur near other words
• Concordance and indexing programs are used widely in literary analysis and
lexicography
• Authors seem to favor using certain words in the context of other words ,
concordance program can find these relationships

Lecture 6-1

Uploaded by

Copyright:

Available Formats

Lecture 6-1

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Lecture 6-1

Uploaded by

Copyright:

Available Formats

Introduction to Language Technologies:

Challenges and Applications

Prof. of Computational Linguistics

2) Information Retrieval (IR)

3) Man-Machine Interface/ Human Computer Interaction (HCI)

❖Corpus creation and documentation E.g., CIIL- LDC- IL Corpora

❖Word Sense Disambiguation (WSD) E.g., Machine Learning Algorithms

❖Natural language Question and Answering (QA) System

❖Optical Character Recognition (OCR)

❖Sentiment Analysis/ Emotional Identification

❖Information Extraction (IE) E.g., Google Search engine

❖Spelling & Grammar checking

❖Automatic Morphological analysis and generation

❖Named Entity Recognition (NER)

❖Parts- of- Speech Tagging / Annotation

❖Speech Synthesis E.g., Speech- To- Speech (STS)

❖Natural Language Dialogue Interfaces to Databases

❖Spoken dialogue system

❖Speech corpus creation and development

❖Voice dialing system

❖Automatic phonetic dictionary

Spoken Speech Written Machine Written Speech Spoken

• e.g. dictation systems

• e.g. automated call-centers.

Constructing linguistic outputs from non-linguistic inputs; the NLG

“I'm sorry, Dave, I'm afraid I can't do that”

• A concordance tells which words occur near other words

You might also like