Businesses are turning to digital assistants to scale communications and meet customer needs. But to do that, these tools need to be better able to replicate the human experience than voice assistants of the past.
On this latest episode of Haptic Feedback, Jamie Young, Director of Product Design with Shockoe, spoke with Nathan Stratton, Founder and CTO of Vocinity, a company providing conversational voice and video interfaces for businesses. Together they discuss the past, present, and future of voice and video technology.
To start, Stratton discussed why he saw the need for Vocinity. “After calling Amazon to route a package I expected to be greeted by Alexa and that they would be using conversational natural language voice interfaces.” When this wasn’t the case, Stratton set out to improve the space.
Vocinity sees opportunities for voice and video avatars in a number of areas, and works closely in consumer facing industries like retail and healthcare. And during the pandemic, demand grew “One example is our work with an exercise bike manufacturer. Because of the COVID-19 pandemic they needed to answer questions about the product differently, so we built a bot to interact with customers.”
Young and Stratton discussed the exciting opportunities for voice and video in the future. But first, Stratton shared an important hurdle for voice and video, the uncanny valley. “That’s a term that means that the closer you get to human-like, the more the mind picks up that it isn’t real. We countered that by using human actors in pre-recorded video that can be adapted for each interaction.”
And while speech to text is a newer technology, as Young pointed out “speech to text and text to speech play a large part in accessibility especially for visually impaired.” While Stratton agreed he highlighted that advancements still can take place in speech to text to improve its recognition of ESL speakers.
With voice and video advancements happening all the time, Vocinity continues to innovate on their technology, with the goal of someday developing lifelike models. These may be closer than you think!