Hand-held text-to-speech device for the non-vocal disabled

Area of Science:

• Assistive Technology
• Speech Synthesis
• Biomedical Engineering

Background:

• Non-vocally disabled individuals require effective communication aids.
• Existing speech aids may lack portability or advanced features.
• Rule-based speech synthesis presents challenges in intelligibility and naturalness.

Purpose of the Study:

• To develop a hand-held, battery-powered synthetic speech aid for the non-vocally disabled.
• To improve the quality and intelligibility of synthesized speech from a rule-based system.
• To incorporate user-friendly features for portability and ease of use.

Main Methods:

• A text-to-speech (TTS) system was designed using 349 letter-to-sound rules and continuity/intonation/stress algorithms.
• The system utilizes a single-chip series formant speech synthesizer and CMOS components for portability.
• Power-saving features and a unique one-handed keyboard were implemented for extended battery life and ease of operation.
• A two-line liquid crystal display (LCD) with editing functions facilitates message composition and silent communication.

Main Results:

• A functional hand-held synthetic speech aid was constructed.
• The device converts user-keyed text into audible speech.
• Efforts were made to enhance speech quality and intelligibility within the constraints of the synthesizer and portable design.

Conclusions:

• The developed device offers a portable and accessible solution for non-vocally disabled individuals.
• The system demonstrates advancements in rule-based speech synthesis for assistive technology.
• Further improvements in speech quality may be limited by current synthesizer technology and system constraints.