COST250 Speaker Recognition Reference System

9. File format description

9.1 Experiment specification

The main purpose of a reference system is to use it with data and experiment specifications that maybe nobody else has used it with. The experiment specification file is what defines how an experiment is run on a particular data set, and is therefore central to the system.

The purpose of an experiment specification file is to define a set of operations to perform. An operation is either an enrollment operation or a verification operation (in the future identification operations will also be included). One operation is defined on exactly one line in the file.

An enrollment operation involves training a speaker model for a certain identity from a set of files. A line that defines an enrollment operation has the following format:

     enroll identity file1 ... fileN 
A verification operation involves a speaker claiming an identity, using a set of files to support the claim. A line that defines a verification operation has the following format:
     speaker identity file1 ... fileN 
speaker and identity are strings like M010 or F031 where M indicates a male and F a female speaker. filei is a file tag rather than a complete filename. The file tag will be prefixed by the name of the database's base directory and suffixed by a file name extension. For Polycost, the file tag has the format
     speaker/session/filename
where filename is a string like DIG05.

Empty lines; matching angle brackets (<...>) and all characters in between; and a hash mark (#) with all the following characters on the line are ignored and can be used for comments in the file. Note that HTML/XML tags are text within angle brackets, and therefore experiment specification files can be full-featured HTML/XML files.