Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Voice-To-Digital Conversion Of Speech For Bandwidth Compression

IP.com Disclosure Number: IPCOM000097333D
Original Publication Date: 1962-Oct-01
Included in the Prior Art Database: 2005-Mar-07
Document File: 2 page(s) / 36K

Publishing Venue

IBM

Related People

Dersch, WC: AUTHOR

Abstract

This system compresses speech frequencies at the bandwidth extremities so as to accommodate the speech wave to narrow bandwidth transmission. The devices employed are notched-filters at the transmitter and voice recognition devices with a digital encoder at the receiver.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 99% of the total text.

Page 1 of 2

Voice-To-Digital Conversion Of Speech For Bandwidth Compression

This system compresses speech frequencies at the bandwidth extremities so as to accommodate the speech wave to narrow bandwidth transmission. The devices employed are notched-filters at the transmitter and voice recognition devices with a digital encoder at the receiver.

The audio speechwave is presented to microphone M and then sent in parallel to filters F and a pair of friction detectors, friction weak detector F(W)D and friction strong detector F(S)D. Friction is synonymous here with fricative. Filter F includes two notched filters. F1 removes the 2.8 KC portion of the raw wave. F(2) deletes the 3.3 KC portion. These are reinserted in the wave as transmitted from transmission point T. In place of the notched portions, a coded signal is substituted and represents the presence of frictioning.

This is produced by oscillators responding to the detection by either the frictioning detectors F(W)D and F(S)D of weak and strong frictioning, respectively. The ultimate effect is that instead of the transmitted wave having the frictioning extremities which are unsuitable for transmission, a pair of digital signals representative of the occurrence of these frictioning components is inserted. Such is in the notched out portions of the bandwidth and then decoded to present this speech information to the speech recognition machine at the receiving point R.

1

Page 2 of 2

2

[This page contains 2 pictures or other...