CallMiner Partners with SpeechPro to Separate Speakers in Call Center Recordings

SpeechPro, a provider of voice biometrics solutions, and CallMiner, a provider of speech analytics solutions, have joined forces to bring speaker separation capabilities to CallMiner’s speech analytics product suite.

Through this partnership, CallMiner will now offer SpeechPro's proprietary speaker separation technology as an add-on to its Eureka call center platform. This new technology, the hallmark of SpeechPro's VoiceGrid product series, separates speakers in a recorded conversation, improving the depth and detail of speech analysis available for call centers.

"CallMiner's customers will get an even better analysis of their contact center recordings," says Alexey Khitrov, president of SpeechPro.

The technology uses SpeechPro's voice biometrics to isolate each individual voice in the recordings. "It looks at each part of the recordings in increments and identifies to whom each voice belongs," Khitrov explains. "Within the VoiceGrid, it can separate out the speech of all the various speakers."

According to Khitrov, most legacy call recording equipment keeps the agent and the customer in the same channel. "That makes it tough for the analytics, because you can't always tell who is speaking," he says.

"In the contact center environment, the typical conversation is between two people: a customer and an agent," Khitrov says. "But we can work with up to five speakers on the same channel at once."

That could be helpful for when supervisors jump on a call or calls get escalated or transferred to other agents, Khitrov points out.

"We see a lot of our customers stuck with legacy recording equipment and contact center infrastructure that offers no speaker-separated audio," said Erik Strand, vice president of product at CallMiner, in a statement. "Analyzing non-separated audio is possible, but it requires significant additional effort and doesn't allow for speaker-based searches and categorization. SpeechPro's speaker separation capabilities allow our customers to take advantage of the benefits of speaker-separated audio and reduced analytical effort without having to change any of their current processes."

The partnership, Khitrov says, makes a lot of sense for both companies.

Khitrov calls CallMiner, "an innovative company that is constantly on the lookout for new technologies that help expand the capabilities of their speech analytics offerings."

Prior to launching the technology and making it available to companies like CallMiner, Khitrov says SpeechPro conducted extensive research and development and product testing.

As a result, "the depth and accuracy now available through the Eureka solution will ensure that we are delivering more value to call center customers,” he adds.

Earlier this year, SpeechPro announced a similar deal with Aspect to bring its VoiceKey biometrics capabilities to Apect's Prophecy IVR platform.

---