Voice Search Sees Business Boost
Observe.AI technology also automatically redacts private data in audio recordings and the corresponding transcripts so that it can’t be funneled into reports on contact center performance and suggestions for improvements.
Without this form of technology, companies can typically only monitor and analyze between 1 percent and 2 percent of all calls; with it, they can work with 100 percent of calls without taking up a huge amount of computing resources.
Oracle recently rolled out a major update for its Oracle Digital Assistant, which now includes native multilingual natural language understanding and customizable vocabulary. The Oracle Digital Assistant, which debuted in 2016, gained voice capabilities last fall.
With the latest update, the AI can be taught with data in a client’s native language and respond in that language, eliminating the need to translate everything into English and back again before the voice assistant can be invoked.
Oracle also integrated its own software architecture into the learning models to improve the performance of the AI when it comes to differentiating and understanding sentences that might seem similar or that use metaphors, intuiting what is being asked of it using context.
“Chatbots and conversational AI are quickly becoming integral tools for enterprise communication and information sharing, in addition to automating traditionally manual tasks,” said Suhas Uliyar, Oracle’s vice president of product management for Digital Assistant, AI, mobile, and integration, in a statement. “With the new updates to Oracle Digital Assistant, we are delivering the innovative features users are seeking, such as multilingual capabilities, to further weave digital assistants into the fabric of the enterprise.”
The need for voice search also becomes more important today as many companies have adopted technologies like Zoom Video Communications’ platform and Microsoft Teams. Companies like Verint recently launched a Microsoft Teams recording integration to help businesses capture, retain, analyze, and retrieve all communications from Microsoft Teams’ calling and meeting scenarios,
“Amid the current COVID-19 crisis, we have seen increasing demand from financial services, healthcare, government, and other regulated organizations to enable continuity of business communications and critical workflows while adhering to the highest compliance standards when working remotely,” said John Bourne, senior vice president of global channels and alliances at Verint, in a statement. “Relying on this secure integration, businesses can adopt or transition to Microsoft Teams in a compliant manner, address the need for undisrupted collaboration, and efficiently support regulatory policies for record keeping, communication surveillance, data protection, reporting, and e-discovery.”
Another factor behind the increased interest in enterprise voice search is the need to meet increasingly complex regulations around the collection and use of voice data. Those regulations, which can differ from country to country and even from one U.S. state to another, cover issues like disclosures—for notifying consumers that calls might be recorded and how those recordings could be used—and security.
Current voice-enabled systems can automatically locate and call up the appropriate disclosures based on unique details of the agent-customer conversations, such as the customers’ location or the type of industry involved. Once the requirements are identified, companies can have agents read disclosure scripts or play automated recordings of the disclosures.
Contact centers must also ensure the data security of all recordings. The same voice search capabilities enable contact centers to locate and replay stored recordings for a range of applications, including dispute resolution, compliance, quality management, and event reconstruction, according to Speechmatics’ Firth.
Challenges in Adoption
Despite the advantages of voice search, the technology still has many adoption challenges to overcome. Two-thirds of companies in the AppDynamics survey that have not already invested in voice technology say the primary factor is fear of a poor user experience. Eighty-eight percent said they had experienced an error rate above 6 percent, while 35 percent said they experienced an error rate of more than 20 percent.
Additionally, 57 percent cited voice recognition accuracy as the major industry challenge. Another 54 percent cited difficulties making sure the technology was worth the cost as a road block to adoption.
Nonetheless, until the challenges and relative novelty of the technology on the enterprise side are overcome, adoption will continue to be limited to leading-edge, large enterprises, many experts fear.
Phillip Britt is a freelance writer based in the Chicago area. He can be reached at spenterprises@wowway.com.