Browse Prior Art Database

Speech Normalizing Technique

IP.com Disclosure Number: IPCOM000098264D
Original Publication Date: 1960-Jun-01
Included in the Prior Art Database: 2005-Mar-07
Document File: 2 page(s) / 21K

Publishing Venue

IBM

Related People

Wortzman, D: AUTHOR

Abstract

The arrangement obtains a digital representation of an input waveform generated by speech. If the same word is spoken by two people, the waveform may be similar but may differ in amplitude or loudness. To obtain the same digital output from an analog-to-digital converter, means must be provided for normalizing the input waveform.

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 100% of the total text.

Page 1 of 2

Speech Normalizing Technique

The arrangement obtains a digital representation of an input waveform generated by speech. If the same word is spoken by two people, the waveform may be similar but may differ in amplitude or loudness. To obtain the same digital output from an analog-to-digital converter, means must be provided for normalizing the input waveform.

In the circuit the input wave is applied at 1. The wave is rectified in 2 and filtered at 3 before application to the analog-to-digital converter 4. The rectified and filtered wave is used as the reference voltage 5 in the converter. The original wave is applied to the converter at 6. The digital output of 4 is a representation of the ratio between the input wave amplitude and the rectified and filtered wave.

If the same spoken word is applied at 1 at twice the amplitude, the digital output is the same. The ratio between the input wave and the reference voltage, now twice as great as before, is the same as the wave at one-half the amplitude.

1

Page 2 of 2

2

[This page contains 1 picture or other non-text object]