Real-time speech visualization system: Kannon - Applying auditory characteristics

We have been developing a real time speech-displaying system called “KanNon” which helps deaf person to understand speaker's speech contents. We designed the KanNon system to display a sound spectrogram, pitch frequency and loudness of speech as well as characters by speech-recognition system as real-time scrolling image. For the purpose of displaying formant patterns clearly with high accuracy, we applied Burg method combining with the minimum cross-entropy (Burg-MCE) method, and human auditory characteristics such as an equal loudness preemphasis and mel-scale frequency to the sound spectrogram. Finally, we show more effective display for the spectrogram reading in the KanNon system.