AI Turns Voice Calls Inside Out
Applying artificial intelligence to speech will forever change the way we think about customer conversations.
Alexander Graham Bell's inaugural voice transmission lives in technology annals as one of the biggest disruptors of its age -- if not of all times. Fast-forward 140-some years, and oddly enough the phone call has once again become a disruptive force, so to speak.
By that I mean having to place a phone call for customer service these days is quickly rising to the status of disruptor -- and not in a positive way. Consumers want seamless connectivity and quick help afforded via self-service Web portals and mobile apps. Needing to escalate to a voice call? That's not good. And while being able to click to call from within a site or app is less troublesome than having to pick up a phone and punch in numbers on a dial pad, phoning into a contact center today often comes cloaked in irritation.
Yet voice -- but not voice calls -- is very much the future of the industry, as Joseph Williams, UC analyst and consultant, noted at the UCStrategies' recent annual summit on business communications.
The changing customer experience dynamic places the voice-oriented contact center in the bull's-eye for intervention and transformation, hence why we've seen so much conversation and activity around the role of artificial intelligence (AI) in customer service operations. Certainly the big players in the contact center and CRM are going all in -- think Genesys with Kate and Saleforce with Einstein, for example -- but this is, as Williams said, "where the money is" for startups, too.
A year ago, Williams said he predicted that within five years AI would make the things in the industry amazingly different. "And I was wrong," he confessed. Why? Because the disruption, driven by innovators, entrepreneurs, and investors, is already happening, he said.
During his presentation, Williams highlighted a few examples. Stealth-mode startup Tethr transcribes the audio recordings of every single contact center call and runs analytics to deliver three classes of insights: performance feedback for customer service reps, real-time product trends for engineering teams, and overall assessment for an executive dashboard. Another startup, Defined Crowd, essentially does the same but adds in sentiment analysis, and Celebri AI offers analytics on the customer journey, including those call center interactions.
Such capabilities are popular with digital native companies like Uber, HomeAway, and Airbnb, Williams said. These sorts of companies "don't have a lot of dial-tone presence in the world, but when somebody's calling they're calling because they're upset about something and need resolution fast."
Moving forward, companies -- digital native or otherwise -- are increasingly going to recognize the need for "conversational intelligence," Williams said. He defines conversational intelligence as the "combination of technologies spanning voice and text-based communications used in AI systems to analyze, classify, understand, enhance, and automate conversations." Example capabilities include:
- Voice -- voice to text, speaker identification, dialect recognition, voice tone analysis, mixed language
- Text -- entity extraction, sentiment analysis, grammar inference, data redaction, dialogue matching
- Reasoning -- call intent inference, automated diagnosis, sales insights, chat bot engines, knowledge index
AI is not magic, Williams said. But, when combined with advanced speech technologies, it sure may seem that way compared to the insight -- or lack thereof -- previously available on the customer experience. With his famous words, "Mr. Watson, come here, I want you," Bell ushered in a communications renaissance circa the late 1800s. AI-embued voice technologies are doing the same now.
Learn more about Speech Technologies at Enterprise Connect 2018, March 12 to 15, in Orlando, Fla. Register now using the code NOJITTER to save an additional $200 off the Advance Rate or get a free Expo Plus pass.