Outline

Any content on Language Engineering (LE) Applications will include (1) generic technologies, (2) standard applications, (3) systems design, and (4) tools and platforms. Generic technologies include the basic methods needed to process speech and text. Standard applications are the typical areas where working systems can be applied. These do not change much over the years although their specific manifestations do. Systems design is where everything comes together by learning how to design prototype application systems incorporating generic technologies. Tools and platforms focus on the basic tools which everyone uses but also larger comprehensive platforms now available for building more integrated systems. Group work and project-based education are important here due to the need for integrating the processing of different phenomena such as signals and symbols, but also multimodality.

Topics

• Generic technologies
  • speech recognition and synthesis/TTS
  • speaker identification/verification
  • text processing (preprocessing + parsing) and generation
  • text analysis and markup (SGML, XML, TEI)
  • integration of speech and text processing (signals <-> symbols <-> symbols)
  • multimodal integration (e.g. with visual input)
  • data collection (ELRA, LDC), requirements, evaluation and training
• Standard applications (including design issues, speaking environment, human factors, and examples)
  • Command-and-control
  • Intelligent MultiMedia (MultiModal) systems
  • Spoken dialogue systems
  • Information Retrieval (IR) and Extraction (IE)
  • Dictation
  • Machine Translation (MT)
  • Computer Aided Language Learning (CALL)
• Systems design

• Requirements, development, evaluation and testing
• Empirical methods (databases, knowledge elicitation)
• Architectures & communications systems
• Standards: e.g. EAGLES

• Tools and Platforms

• tools, e.g. Waves, Matlab, Entropic HTK
• evaluation and usage of a typical platform including speech and NLP (typically as groupwork or project-based building of prototype application systems)

References

1. [Ber]: Bernen, Niels Ole, Hans Dybkjaer and Laila Dybkjaer (1998) Designing interactive speech systems: from first ideas to user testing, New York: Springer-Verlag.
2. [Cole]: Cole, Ronald A., Joseph Mariani, Hans Uszkoreit, Annie Zaenen, Victor Zue, Giovanni Varile, Antonio Zampolli (1995) Survey of the state of the art in human language technology, (www.cse.org.edu/CSLU/HLTsurvey.html), (www.elsnet.org/publications/hlt).
3. [Del]: Deller, J., J. Proakis and J. Hansen (1993) Discrete-time processing of speech signals, New York, US: MacMillan Publishing Company.
4. [Gibb]: Gibbon, Dafydd, Roger Moore and Richard Winski (Eds.) (1997) Handbook of standards and resources for spoken language systems, Spoken language system and corpus design, Spoken language characterisation, Spoken language system assessment, Spoken language reference manuals, Berlin, Germany: Mouton de Gruyter.
5. [IIR]: Inventory of Internet Resources, SOCRATES Thematic Network in speech communication sciences, http://www.phon.ucl.ac.uk/resource/cal-top.htm
6. [Jel]: Jelinek, F. (1998) Statistical Method for Speech Recognition, MIT Press, Cambridge Massachusetts.
7. [Mar]: Judith Markovitch (1996) Using Speech Recognition, Prentice Hall.
8. [McKev]: Mc Kevitt, Paul (Ed.) (1995/1996) Integration of Natural Language and Vision Processing (Vols. I-IV), Dordrecht, The Netherlands: Kluwer-Academic Publishers.
9. [Young]: Young, Steve and Gerrit Bloothooft (1997) Corpus-based methods in language and speech processing, Dordrecht, The Netherlands: Kluwer-Academic Publishers.