EAGLES Handbook on Spoken Language Systems DRAFT --- WORK IN PROGRESS

EAGLES DOCUMENT SLWG Phase 2, V1

Version of 18 May 1995

EAGLES ContentsSLWG--Handbook Phase 2, V1 FOR EAGLES RESTRICTED USEContents--DRAFT 18 May 1995

• Contents
• Introduction
• System design
  • Introduction
  • System capability profile versus application requirement profile
  • Technical features versus system capabilities
  • Speech recognition systems
  • Signal module
  • Speaker verification/identification
  • Speech synthesis
  • Interactive voice systems
  • Software aspects
  • Hardware aspects
  • Multi-linguality aspects
  • Conclusions
• Corpus design
  • Introduction
  • Seven main differences between collections of written NL- and spoken SL-data
  • Applications of spoken language corpora
  • Specification of the linguistic content
  • Specification of number and type of speakers
• Corpus collection
  • Introduction
  • Data collection dimensions
  • Monitoring
  • Procedures
• Corpus representation
  • Introduction
  • Transcription of spoken language corpora
  • Storage and design of the data base
• Spoken Language Lexica
  • Introduction
  • What is a spoken language lexicon?
  • Types of lexical information in spoken language lexica
  • Lexical surface information
  • Morphological information
  • Grammatical information
  • Lexical content information
  • Lexicon structure
  • Lexical knowledge acquisition for spoken language
  • Outlook
• Language models
• Dialogue
• Physical characterisation
• Assessment methodologies and experimental design
  • Introduction
  • Statistical and experimental procedures for analysing data corpora
  • Assessing recognisers
  • Assessing speaker verification and recognition systems
  • Interactive Dialogue systems
  • Appendix 1
  • Appendix 2
• Recognition assessment
• Assessment of speaker verification systems
  • Presentation
  • A taxonomy of speaker recognition systems
  • Influencing factors
  • Scoring Procedures
  • Complementary assessment tools
  • Typology and assessment of applications
  • Forensic applications
  • Conclusions
  • References
• Synthesis assessment
  • Introduction
  • Towards a taxonomy of assessment tasks and techniques
  • Methodology
  • Black box approach
  • Glass box approach
  • Further developments in speech output testing
  • List of recommendations
  • Appendix 1: Summary of test descriptions
  • Appendix 2: Evaluation of visual aspects of speech output
• Spoken language interactive systems
• Tools
  • Introduction
  • Signal theory
  • Speech research
  • Computer hardware and software
  • Conclusion
  • Appendix: Useful anonymous ftp sites
• References
• Computer readable alphabets
• SAMPA computer readable phonetic alphabet
  • SAMPA
• SAM label file formats
  • SAM label file formats
• SAM recording protocols
  • SAM recording protocols
• SAM software tools
  • DKISALA (Interactive Semi-Automatic Labelling Software)
  • ELSA (ESPRIT Labelling System Assessment software)
  • EUROPEC (European Program d'Enregistrement de Corpus)
  • PTS (Progiciel de Traitement de Signal)
  • RESAM
  • SAMITPRO (SAM Iterative Proportional Fitting)
  • SAM_REC0 Isolated Word Recogniser
  • SAM_SCOR (Sam Input Assessment Scoring Software)
  • SAM_SLM
  • SAM_SPEX (Speech Parameter Extractor)
  • SAMTRA (SAM TRanscription Analysis)
  • SOAP (Speech Output Assessment Package)
  • PTM (Parametric Test Manager)
• EUROPEC recording tool
• EUROM-1 database overview
• Polyphone project overview
  • Introduction
  • Reusable resources
  • The Dutch POLYPHONE corpus
  • Use of POLYPHONE in application development
  • Requirements for future corpora
  • References
• Overview of speech corpora
  • Overview
  • Criteria for assessment of the situation of Spoken Language Resources
  • Summary of the current situation on a per language basis
  • General conclusions
  • Production costs
• About this document ...