Custom-trained speech-recognition models from Sensory understand the linguistic patterns associated with children’s speech. The models take advantage of Sensory’s TrulyHandsfree phrase-spotting technology and TrulyNatural large-vocabulary continuous-speech recognizer. Developers of apps, toys, and wearables for children can implement accurate voice control with privacy thanks to the Sensory AI-on-the-edge architecture.
To meet the challenge of recognizing children’s speech, compounded by the scarcity of available training data, Sensory collected and analyzed significant amounts of children’s speech to better understand and model the specifics of how children talk. Initial testing of this corpus of spontaneous speech resulted in a reduction of the word error rate by as much as 33% compared to an adult speech-recognition model.
Developers can access the child speech models, as well as adult speech models, within Sensory’s VoiceHub developer portal. VoiceHub enables direct export to many supported DSP and microcontroller formats from a wide network of global partners, including the recently added Taiwanese chip maker Generalplus Technology.
Find more datasheets on products like this one at Datasheets.com, searchable by category, part #, description, manufacturer, and more.