Oracle Gives Voice Assistant Multilingual Learning Capability
Oracle rolled out a major update for its voice assistant this week. The Oracle Digital Assistant now includes native multilingual natural language understanding and customizable vocabulary. The updates may help Oracle stand out as a voice-enabled enterprise service as new rivals appear.
The Oracle Digital Assistant has been around since 2016. At first purely a chatbot, Oracle added voice capability to the assistant last autumn. Vocal communications opened up new dimensions of interaction for Oracle’s clients to connect with sales and other software, streamlining a lot of administrative duties along the way. Now, that communication can be run in a variety of languages using the native multilingual NLU feature. The AI can be taught with data in a client’s native language and respond in whatever language best suits the client at that time. That means there’s no longer a need for a company using Oracle’s software to translate everything into English and back again to be able to use the voice assistant. 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 may seem similar, or that use metaphor, intuiting what it is being asked using context.
“Chatbots and conversational AI are quickly becoming integral tools for enterprise communication and information sharing, in addition to automating traditionally manual tasks,” Oracle’s AI and Digital Assistant vice president Suhas Uliyar said 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.”
Oracle’s upgrades to its voice assistant match the ambitions of many companies looking to integrate AI and voice technology into their operations. Oracle does have to face off against the enterprise ambitions of voice assistant developers like Amazon and Goole. They have been upping the capacity of Alexa and Google Assistant respectively to handle more business needs. Microsoft sees so much potential in the enterprise side of voice that it has completely shifted its Cortana voice assistant away from consumers. Oracle does have some advantages by building voice capabilities into existing business tech as opposed to adding enterprise abilities to a voice platform. The company has a lot more data about business language and expressions that it can apply to the AI, for one, and Oracle can point to the fact that data is kept in its own cloud system as a security bonus for companies that are wary of Amazon or Google’s security.
That doesn’t mean Oracle’s voice assistant has an assured future, especially in the current tumultuous economy. Enterprise AI is still risky, as Australian virtual assistant developer Flamingo AI learned after failing to raise enough capital led to its sale for $100 Australian dollars after seven years of growth. And, while Salesforce as a company is going strong, its Einstein voice assistant program was just shut down, and its developers moved into other applications.