RSC-164i

The RSC-164i, from the Interactive Speech™ family of products, is a low-cost microcontroller designed for use in consumer electronics. The RSC-164i combines an 8-bit microcontroller with high-quality speaker-independent and speaker-dependent speech recognition, speaker verification, and speech synthesis. Products can use one or all of the RSC-164i features in a single application.

The RSC-164i employs a sophisticated neural network that learns to classify sound data. On-chip speech recognition algorithms reach an accuracy of greater than 96% for speaker-independent recognition and greater than 99% for speaker-dependent recognition. Sensory's neural network approach (patent pending) eliminates the need for expensive signal processing or extensive RAM storage.

The highly-integrated nature of the chip reduces external parts count. A complete system may be built with few additional parts other than a battery, speaker, microphone, and audio input support circuitry. The RSC-164i is similar to the RSC-164, but does not support a parallel memory interface. All program, synthesis, recognition, and verification code must fit in the on-chip ROM.