iii
Preface
The System Requirements Specification is one of the critical documents since it identifies
the aims and goals of the project. This is important to the success of the project since it
clearly outlines what the development team must achieve in order to classify the project
to be complete.
The project’s aim is to develop a telephonic speaker recognition system. The first step is
to create a speaker database containing digitized speech recordings of people of both
sexes which are to be identified. These speech recordings should contain both telephonic
as well as non telephonic data. The speaker verification will pursue text-independent
method in which, speaker models capture characteristics of somebody’s speech which
show up irrespective of what one is saying, and text dependent method in which, the
verification is based on what the speaker is saying. The text dependant and independent
methods will be applied on telephonic and non- telephonic speech data.
Later, comparisons would be made among the speaker recognition results acquired from
telephonic data, which will count the handset variability as well as channel distortion, and
the non telephonic speech. Speaker Recognition via lip motion may be carried out at later
stage.
Speaker Recognition makes it possible to use speech to verify identify and control access
to services such as computer-human interaction, voice dialing, banking by telephone,
voice activated data entry in medical or dark room, telephone shopping, voice mail,
security control for confidential information areas.
docsity.com
