Continuing to mine the speech recognition niche in many embedded consumer and industrial applications for all it is worth, Sensory, Inc. released the newest member of its RSC-4x family, the RSC-464, supported by a suite of development tools.
According to Todd Mozer, Sensory's CEO, the RSC-464 is a powerful general-purpose microcontroller inside a speech recognition system-on-chip that includes
16-bit ADC, DAC, digital filter unit, math unit, 4K RAM, 64K ROM, output
amplification, timers, comparators and more.
While it can operate as a standalone speech recognition MCU, he said it can also be used as a slave for speech recognition, voice biometrics, and speech and music playback, making it easy to add to existing product designs with minimal changes to the primary controller software.
To support speech recognition application development, the company also provides its FluentChip firmware library which includes speaker independent (SI)
recognition, speaker dependent (SD) recognition, speaker verification
(voice password) biometrics, voice record, speech compression/playback and
MIDI-like music synthesis.
FluentChip 2.0 offers new options for combining SI and SD in a single command set as well as new algorithms which allow the RSC-464 to control mouth movements of animated dolls so the movements appear naturally in time with speech.
Additional tools include Quick Synthesis, which allows fast development of
speech output files by recording and compressing a file with the touch of
a button, and also supports easy scoring of MIDI-like music. A
fully-integrated IDE for software development is provided by Phyton, Inc.,
including assembler, linker, debugger and file management, as well as an
optional C compiler to accelerate coding.
The RSC- 464 IC sells for under $1.10 in large volumes.
Santa Clara, Ca.