Voice and Speech Systems

Products > Computerized Articulation Therapy

Vagmi Articulation Therapy
a comprehensive system for articulation therapy

Vagmi Articulation Therapy is easy to operate and user friendly software package. It requires no programming skills.

Contents of this Page: (A) Free Download (B) Introduction (C) Articulation Therapy Groups

A. Free Download
Power Point Presentation of Demo:
Instructions:Right click on the link and 'Save Target As:

Vagmi_Therapy_Articulation_DEMO.pps

B. Introduction

Utility of Vagmi Articulation Therapy:

Hard of hearing and spastics

Persons with hyper or hype nasal problems

Vocal musicians singing at high pitch producing a nasal voice

What is Articulation Therapy?

Mis-articulation is a common problem even amongst normals. Children often mis-pronounce fricatives such as 's', 'sh' etc. 'Deaf needn't be dumb'. They have normal control over articulation. Problem of hard of hearing is much more complex. As they rely on lip-reading, they mis-articulate sounds 'b', 'p' and 'm'. Voice (g,d,b) Vs Unvoice (k,t,p) distinction is another major problem. Stroke patients have to undergo speech therapy to regain speech. Listening to one's own tape-recorded voice or standing in-front of a mirror is not an effective therapy technique.
Vagmi Articulation Therapy is a sophisticated software tool that converts a PC into a therapy equipment. What used to be monotonous and time consuming has now become interesting, effective and time saving. Vagmi Articulation Therapy provides a visual feedback of articulation process and displays the measured parameter in comparison with a model. The articulation of the phoneme can be shown in a Script of a language of user's choice such as English, Hindi, Kannada etc. It has many interesting Games, Puzzles etc with powerful visuals. Stimulating tasks are to be achieved during therapy. The performance is scored. Reinforcement messages appear. Goals can be set in a graded manner.

C. Articulation Therapy Groups

Menus under Vagmi Artiuclation Therapy are divided into two broad groups:

Group:A Vowels and Fricatives

Group B: V/UV (Voiced and Unvoiced) Production

Customer can indent either group A or group B or both.

Download the power point presentation demo to study the details of the each of the menus.
See link at the top of the page under 'A. Free Download'.

Group A: Vowels and Fricatives:

Vowel Therapy:

There are 4 menu options under vowel therapy:

F1 Vs F2 Space: The vowel to be practiced along with gender and age are selected. A graph of Second Formant (F2, y-axis) Vs First Formant (F1, x-axis) is displayed. A rectangular box whose location depends on the goal set appears. For a sustained vowel sound, first two formants are computed and their location is displayed by a dot in F2 Vs F1 sapce. When the vowel is uttered correctly, the dot lies within the rectangular box. Vowel identity is shown in the script of a chosen language. This option also acts as an automatic vowel identification program.
Feature Space: Vowel sounds are categorized as 'High/Low' and 'Front/Back' based on the tongue profile. 'High/Low', 'Front/Back' are called features. Thus vowel 'aa' is a Low-Back vowel. Vowel 'ee' is a High-Front vowel. Vowel 'A' is a 'Back/Front' vowel. Vowel 'oo' is a High-Back' vowel. Vowel 'O' is 'Mid-Back' vowel. Further, lip rounding is another feature.
Vowel to be practiced is selected. A graph of High/Low (y-axis) Vs Front-Back (x-axis) is displayed. The feature space is divided into rectangular boxes. For a sustained vowel sound, the features are computed. Their location is displayed by a dot in the feature sapce. When the vowel is uttered correctly, the dot lies within the correct rectangular box. Vowel identity is shown in the script of a chosen language. This option also acts as an automatic vowel identification program.

Tongue profile Vowel sounds are categorized based on the tongue profile. A model mid-sagital profile of vocal tract with the tongue profile is shown for the selected vowel. For a sustained vowel sound, the tongue profile is computed. Its location is displayed relative to the model. When the vowel is uttered correctly, a black curve is superposed on the model red tongue profile. Generally it is said that the mapping from acoustic signal to tongue body profile is not unique but depends on gender and age. However, we have used a self-normalizing technique to overcome this problem.

Psuedo-Script In handwriting, an alphabet may be written in different styles. yet, the identity of the alphabet can be identified. For a sustained vowel sound, a psuedo-script is computed and displayed. The psuedo-script for a given sustained sound has common features but differ for different sounds. Thus psuedo-script may be used to discrimitae the production of different sustained sounds.

Fricative Therapy:

There are 4 menu options under fricative therapy:

Train: When a sustained fricative is uttered correctly, a train movs emitting smoke. Stronger the energy of frication, higher the smoke emitted.
Beaker: When a sustained fricative is uttered correctly, water gets filled in a beaker. The time taken to fill the beaker completely and the number of breaks in production of frication are shown. A reinforcement appears after completing the task.
Snake: When a sustained fricative is uttered correctly, a snake rises its hood to meet Mowgli. When the friactive is uttered for the set duration, Mowgli is hyptonized when has eyes turn red.
Sh Vs S: After learning fricative production, a finer distinction of producing 'Sh' or 'S' has to be learnt. When a sustained fricative 'Sh' is uttered correctly, picture of 'sheep' appears. When a sustained fricative 'S' is uttered correctly, a picture of a snake appears. Also, the script in the chosen langauge is displayed.

PC Requirement, Support Hardware and Accessories (Click to open the link)

Go Back