Software Components
SFS has its origins in the software developed for speech research by University College London, Imperial College London and GEC Hirst Research Centre under the Alvey information technology initiative. It was originally divided into two components. The first component dealt with the environment in which speech is analysed, the second with the programs for analysis. The first component is now called the Speech Filing System (or SFS) and comprises data structures, utility programs and programming support for speech analysis. The second component was called SPARBASE and comprised a large number of programs for manipulating speech data in its many representations.
This release of SFS incorporates a large number of the UCL SPARBASE analysis programs as developed and enhanced over the past 8 years. Thus as far as UCL is concerned SPARBASE is no longer a separate entity from SFS. The other partners of the consortium, particularly Imperial College and GEC have analysis programs of their own which are compatible with this release of SFS but are not distributed from UCL.
Documentation
The documentation for the SFS speech analysis programs consists of a folder of manual pages. Each manual page gives reference information for a single program.
The documentation for the Speech Filing System environment consists of a manual in 2 parts:
1. SFS for Users
Contains an introduction to SFS suitable for users of existing software. It outlines how speech data is stored in files, how it may be manipulated by processing programs, and how SFS keeps a track of the processing. An annex contains the software manual pages for SFS programs.
2. SFS for Programmers
Contains introductory material for new programmers of SFS speech processing programs, and examples of the use of each of the SFS library routines with full reference descriptions. An annex contains the software manual pages for the support libraries.
This manual, SFS for Users has five sections. Section 1 gives a simple run-through of the operation of SFS running on a Unix system. Section 2 describes how SFS files are created, how data sets are stored, and how the contents of a file may be investigated. Section 3 looks at SFS utility programs for simple manipulation of data sets in files. Section 4 looks in more detail at the contents and structure of data sets. Section 5 gives a tutorial introduction to making speech measurements using the language SML.
Typographical conventions
Times Roman Normal descriptive material.
Courier
Computer text.Underlined Courier
Command input to computer.Red-lined text: Used to signal parts of the text which may be safely skipped on a first reading of the manual.
Mark Huckvale Department of Phonetics and Linguistics University College London Gower Street London WC1E 6BT United Kingdom SFS@phon.ucl.ac.uk http://www.phon.ucl.ac.uk/