Sensory makes significant upgrades to embedded AI in TrulyHandsfree
Sensory has launched version 6.0 of its TrulyHandsfree software with high-resolution digitization for improved performance and integrated voice biometrics for user authentication and device security.
All processing, including biometric matching, takes place on-device, and the embedded high-resolution voice enrollment and speaker verification (SV) technology allows users to enroll their voice, and set a custom wake word or password, which will the device will only accept from the correct user as recognized by Sensory’s voice biometric technology.
TrulyHandsfree also features advanced deep neural network training for better near- and far-field speech recognition performance, and wake word accuracy improvement of more than 65 percent, according to the announcement. The technology reduces wake word false positives by nearly two-thirds by applying its new high-resolution speech feature and on-device wake word post-qualification, which uses information about wake word events to catch false positives.
“In ideal conditions, an always-listening wake word recognizer should only wake the voice UI when it hears the wake word, and never false fire,” said Todd Mozer, CEO of Sensory. “This is how we designed TrulyHandsfree to operate, and with the upgraded AI technologies in version 6.0, it works exactly as it should even in far less than ideal conditions. TrulyHandsfree has long been the performance benchmark that other voice UI solutions strive to match. With our sixth generation, the bar for efficiency, security, wake word accuracy, and near- and far-field performance has been set much higher, making it the ideal wake word solution for any kind of voice-enabled product design.”
TrulyHandsfree 6.0 can also support multiple wake words in a single implementation, and its improved training algorithms make it more tolerant of different room conditions, user accents, and dialects.