Browse Prior Art Database

Multimedia Vocal Tract Visual Feedback

IP.com Disclosure Number: IPCOM000110859D
Original Publication Date: 1994-Jan-01
Included in the Prior Art Database: 2005-Mar-26
Document File: 4 page(s) / 83K

Publishing Venue

IBM

Related People

Williams, FE: AUTHOR

Abstract

The solution to a number of speech impairments lies in the proper control of the elements of the vocal tract. This disclosure addresses a visual feedback method of the vocal tract. It can be used in many applications, but to name two, foreign language training and the correction of speech impairments.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 52% of the total text.

Multimedia Vocal Tract Visual Feedback

      The solution to a number of speech impairments lies in the
proper control of the elements of the vocal tract.  This disclosure
addresses a visual feedback method of the vocal tract.  It can be
used in many applications, but to name two, foreign language training
and the correction of speech impairments.

The advantages of this approach are:

o   Non-invasive method of detecting vocal tract errors

o   Provides a visual feedback method for their correction

o   Is well within the emerging Digital Signal Processor (DSP)
    multimedia technology

      DSP multimedia techniques are used to provide visual feedback
of vocal tract conditions automatically based upon the audio word
received by the system.

      An available word recognition feature card for PC can be used
to extract identifying parameters from speech.  These parameters can
be used in a table lookup to display a visual picture of the vocal
tract positions which generated that speech sound.  For example, in
Fig. 1, the position of the tongue and shape of the lips are
displayed as they should be on one side of the screen, while the
positions of the same elements are displayed as spoken in the other
half.  "Target" audio can be played on the multimedia computer
speaker.  An IBM product, Speechviewer, provides indirect
information, via time waveforms, spectrums and spectrograms.  This
disclosure provides a much better direct method by showing just what
the vocal tract is...