Sensory is trying to make it easier for organizations to take advantage of its voice recognition and computer vision technologies. To that end, the company has unveiled a new SensoryCloud AI-as-a-Service platform that allows clients to add new utilities with an intuitive API.
The SensoryCloud offers support for a number of voice features, including Speech to Text, Wake Word customization, and Sound and Speaker Identification. The Speech to Text engine can be customized to recognize jargon words that are unique to a specific industry, with an error rate lower than five percent. The Sound Identification system can similarly be trained to recognize specific sounds, while the Wake Word feature helps minimize false activation events.
The new platform also comes with Face Verification capabilities. The SensoryCloud is currently available in a beta form, and Sensory has already indicated that it will add new features to the platform as it evolves over the course of the next year.
According to Sensory, the SensoryCloud was developed to meet the client demand for more flexible cloud solutions. The new offering processes voice and computer vision data in the cloud, a fact that sets it apart from the rest of the Sensory portfolio, which places a greater emphasis on edge processing.
“We have a history of building fast and accurate AI models, and we paired this capability with some of the brightest and freshest minds in the cloud industry,” said Sensory CEO Todd Mozer. “The result is a hybrid cloud platform that uses state-of-the-art AI to address customers’ unique needs for control, flexibility, cost, accuracy, reliability, features, latency, and privacy.”
The SensoryCloud is compatible with multiple platforms and is language agnostic, which simplifies management for end users.
The company itself recently released a new speech recognition model that can understand children’s unique vocabulary and speech patterns.