Top NSA Banner

Aliases:

SRSAD

Technical Challenge:

Provide a program that would offer a pre-process for a number of applications that require speech intervals to be identified in heterogeneous audio files to eliminate manual searching, reduce bandwidth required to transmit such audio files, and the amount of memory needed to store them.

Description:

SRSAD eliminates the need for manual search of speech signals by automatically locating speech intervals in an audio file. It outputs the start and stop times of these intervals relative to the beginning of the file. SRSAD ignores the non-speech portions of the file, thereby drastically reducing forwarding bandwidth and storage requirements.

Demonstration Capability:

README file contains instructions for running examples for Unix systems. Versions of the software compatible with Linux and other operating systems may be obtained from the author.

Potential Commercial Application(s):

This program would provide a pre-process for a number of applications that require speech intervals to be identified in heterogeneous audio files. An example may be automatic location of speech intervals for training speaker or language identification systems.

Patent Status:

Issued - United States Patent Number 6,556,967

Reference Number: 1104

If you are interested in exploring this technology further, please call 443-445-7159 or express your interest in writing to the:

National Security Agency
NSA Technology Transfer Program
9800 Savage Road, Suite 6541
Fort George G. Meade, Maryland 20755-6541