Enabling companies to automatically understand the intonation of the human voice.
PROBLEM AND CHALLENGE
Companies don’t have a reliable way of measuring the human quality of spoken conversations with their customers. Today’s speech technology today is simple speech-to-text, which lacks the nuances found in vocal intonations. Key behaviors such as engagement, interest to purchase or satisfaction are found beyond words.
SOLUTION AND OUTCOME
When developing their solution, the team at OTO, an SRI spin-off, leveraged SenSay Analytics™ technology created by SRI International, to power the real-time voice analysis capabilities of their solution. SenSay Analytics provides OTO with real-time audio feature extraction and real-time classification capabilities, making it possible for the company to analyze the sentiment of customer interactions.
As voice-based applications and personalization technologies continue to surge in popularity, developers need access to improved speech analysis systems that go beyond simply turning speech into written text, or recognizing spoken commands.
To help developers meet those increasingly complex needs, researchers at SRI created SenSay Analytics; a platform that performs real-time speaker-state classification from spoken audio. It doesn’t require any pre-segmentation and can be deployed to cloud environments, personal computers and client-based machines.
SenSay Analytics is built with the intention of continuous monitoring and processing of voice data. Voice commands when driving, interactive voice response (IVR) systems and personal assistant technologies are just a few scenarios where SenSay Analytics can be used.
A notable example of the SenSay Analytics platform in use, is OTO, a spin-off of SRI, that develops a multi-dimension conversational system that uses acoustic language processing to help users understand the context, sentiment and behaviors of a conversation.
By using SenSay Analytics, the OTO team was able to take a proven technology platform and use it to assemble their offerings which are centered around enabling call center operators to automatically gauge caller and agent sentiment, helping to improve customer service.
