The research and development of the technology was directed at a variety of physiological elements that impede the fluidity of speech at various levels and in different forms. The technology was designed to analyze and address all the below listed elements in order to be able to restore speech fluidity:

• Long term/high retention levels
• Calibration to individual specifics
• Development of “Correct & Control” faculties
• Speech Dysfluency Evaluation
• Respiratory Pattern Analysis
• Laryngeal Activity Monitoring
• Voice Recognition and Analysis
• Performance tracking & auto-corrective “Expert System”.
• Motor speech optimization.
• Auto-corrective & diagnostic system with biofeedback in real time.
• Articulatory Muscle Tension Identification.
• Voice Stability Analysis.
• Individualized Performance Summaries & Recommendations.
• Adaptation of respiratory-phonatory coordination

In order to coordinate, test, process and monitor the various components of speech production, a special computerized processing center (the Articulatory-Laringo-Lab) and complementing software were developed by our engineering team. Particular attention was given to precise voice recognition and respiratory pattern analysis systems. During the clinical application the treated individual is connected to, and interacts with the processing center unit. The Articulatory-Laringo-Lab incorporates the following systems:

Electroglottograph
This instrumentation is used to measure and analyze vocal cord activity during speech production. It provides the system with all data of the fluctuations of vocal cords activity. The procedure is non-invasive. Two electrodes attached to a velcro band are placed at the larynx level. The electroglottograph tranduces transneck-impedance via an oscillator, specially designed with an amplitude-stable circuit. Technical settings: Its impedance range is between 20 and 250 ohm with a signal to noise level of 48dB at 100 ohm electrode impedance. The frequency range was set between 3.5 Hz to 10,000 Hz to prevent frequency overshoot.

Respiratory Displacement Sensors
These sensors are integrated into belts placed at the thoracic and abdominal level of the individual Through the use of strain gauges they measure displacement of respiratory activities. The collected data is stored for analysis (The system calculates the mean and standard deviation of the signals). The resistance feedback provides thoracic and abdominal changes via custom dual channel 10 bit Analog to Digital converter. Measurements and analysis are conducted to identify the type of respiratory pattern during speech production: shallow, deep, or normal.

Simultaneous Voice and Respiratory Signals Processing
The signals from the voice and respiratory inputs are processed concurrently in a real time mode. Each of the two functions is processed by a separate multi-tasking software program running simultaneously. This provided constant data with each breathing and voice production.

Automated Digital Monitoring and Calibration systems
Speech is processed by Digital Signal Processing software program that receives a digitized input from voice sampling hardware and software. The DSP software program automatically calibrates itself for each monitored sound production within the set parameter(s) of the training time sample. This program determines the noise level of the signal, and detects the beginning and end of the speech segment in comparison with the parameter(s) of the training sample.

Performance tracking and auto-corrective “Expert System”
This computerized system provides simultaneous auto-corrective, biofeedback in real time. Its auto-diagnostic functions include articulatory muscle tension identification, respiratory pattern analysis, laryngeal activity/vocal monitoring, and voice stability analysis. Essential performance analysis & summary for each exercise are generated. The results of each sound category performed by the individual are stored in its data base. Concurrently it computes and compares the sound productions with the established parameters for a correct speech production. At the completion of each exercise session, customized adjustments are displayed in correlation with the targeted skills for each sound category. The recommendations are displayed in a user-friendly configuration and formulated based on the individual’s performance. The results of the performance summary are stored for later review by the speech-language pathologist, and other clinical applications.