Twenty-five percent of people won’t make a call on their cell phones this week, just another data point that reinforces the perception that voice is diminishing in value. In my numerous conversations with Jay Blazensky, co-founder and chief revenue officer of VoiceBase, I couldn’t help but think of how this perception is totally misguided. Why? His company is where voice meets cloud and machine learning. More specifically, newly launched VoiceBase Insights is smart machine technology, allowing predictive insights into business or any other field in which voice or video is used.
The company’s cloud-based voice indexing and discovery platform serves multiple markets. At a base level there is machine translation, and then up the food chain is keyword analytics. Predictive insights is where most of the excitement lies. After a training period of a few thousand calls, the system can do some very helpful things like determine which customers are upset, which are first time callers, who is calling to cancel an appointment, who needs directions, which agents are rude, and who is about to churn.
This is the sort of data that organizations typically don’t have access to today. Going forward, however, a company can see if its customer satisfaction rates via phone are improving or worsening on a daily basis.
The system works in a very interesting way. Fifteen minutes is normal priority to get results. To make this a reality, files are chopped into one-minute pieces. Servers are then spooled up in the cloud and directed to process the file segment. The result is then stitched back together allowing a three-minute and five-hour recording to be processed in the same amount of time.
In addition, Jay explained that archives can be processed more slowly to save money – explaining that this sort of flexibility is his company’s secret sauce, which he believes makes it disruptively competitive. Transcription costs are one cent per minute and analytics costs two cents per minute.
It isn’t difficult to understand why he is so excited about this company as he explains how business processes can now be automated thanks to the ability to determine the type of call a customer has made. Once determined, companies can provide offers, win-back promotions, and a variety of other responses to predetermined behavior.
The technology can also be used to figure out how good sales leads are. For example, it can be used to do A/B testing between different lead-generation vendors.
The bottom line here is that even if people are using their phones less, the calls being made to businesses are likely more important and urgent than ever. In addition, many companies have massive archives of audio and video content, which can now be inexpensively transcribed and analyzed. The concept of big voice is certainly an intriguing one – allowing for the power of big data to be applied to one of the first forms of communications ever used by humans. With VoiceBase, the power of voice can be applied to really automate and streamline how business operates.
Edited by Stefania Viscusi