Speech Technology Magazine

 

Siri, Meet Nina

By Michele Masterson - Posted Sep 10, 2012
Page1 of 1
Bookmark and Share

Nuance Communications in early August launched Nina, a multimodal virtual assistant for mobile customer service apps that allows companies to add speech-based virtual assistant capabilities to their existing iOS and Android mobile apps, enhancing the self-service experience for customers.

Nina combines Nuance speech recognition, text-to-speech (TTS), voice biometrics, and natural language understanding (NLU) technology hosted in the cloud to deliver an interactive user experience that not only understands what is said, but also can identify who is saying it.

"We're really excited about it," says Robert Gary, president and general manager of the Mobile Care Solutions unit at Nuance. "This leverages a lot of the core technologies that Nuance has had in the marketplace for years, and we've added some new capabilities to capitalize on the mobile opportunity and rich dialoguing capability."

Gary says that although Siri has driven awareness of the power of speech in a "mobile metaphor," in the application, Nina and Siri are very different.

"We're focused on a very vertical enterprise space," he says. "It's also being delivered as an enablement tool, not a complete experience and complete solution. Other than that they both use speech, natural language, and dialogue, the focus is on the experience front."

Bill Meisel, president of TMA Associates, agrees. "Nina is much more powerful than Siri within its intended domain. It doesn't require that one use speech," he said in a blog post. "The natural language understanding component can be very intelligent, because it is designed specifically for the customer service environment."

He also notes that Nina is not intended as a personal assistant, pointing out that Nuance's Dragon Go! mobile app is more of a direct competitor to Siri. "If you ask Nina for a weather report, she will be stymied," he wrote.

Nina can authenticate by validating identities using unique voice prints; navigate by capturing intent into a single utterance and getting users precisely where they need to go to get answers; transact by extracting content, context, and meaning from a request to get fast results in fewer steps; and educate users about the product's capabilities.

"Education is important," Gary says. "If there's one thing we've learned in the years of IVR, it's that we need to help people understand what they can do with the technology and what they can say."

Nina also provides an open software development kit (SDK) to support the rapid integration of virtual assistant capabilities into existing mobile applications. In addition, Nina allows organizations to brand their own virtual assistant persona, including the visual appearance and implementation of optional custom TTS voices.

"We've really focused to enable a very flexible design canvas," Gary says. "Enterprise customers can really innovate, even within a product that their competitors may have, because Nina enables a lot of unique design and implementation possibilities. Although we ship with capabilities that come out of the box, it's really open and allows them to expand upon how they want to use it."

The Nina Virtual Assistant for customer service is composed of the following:

  • Nina Virtual Assistant Persona, a reusable reference user interface: Nina offers a complete virtual assistant persona, which developers can leverage for their apps, or use the available source code to create custom personas, including changing visual elements such as being awake, asleep, listening, processing, or answering requests. Users can choose from existing voices or develop custom TTS voices.
  • Multimodal interface: Nina has been designed specifically for mobile app front ends to support inputs via speech, tap, and text, letting users choose their interaction style and move within transactions between multimodal talk, type, and tap.
  • Nina Virtual Assistant SDK: To enable the integration of virtual assistant capabilities into mobile apps for Apple iOS and Android, the Nina Virtual Assistant SDK has three components. Nina Core APIs provide access to the core cloud services, such as speech recognition, TTS, and NLU. Nina Virtual Assistant APIs allow mobile app developers to customize the persona and control all modes of input, including speech recognition, TTS, and touch. Nina Reference Designs—the source code of Nina Virtual Assistant apps and functions, including the Nina Banking Assistant—deliver predesigned templates and tasks for store location, bill pay, account information, and more than 200 other banking-related queries. Developers can leverage the reference designs to develop their own virtual assistant capabilities for travel, insurance, retail, government, and more.
  • Nina Virtual Assistant Cloud: Nuance's speech recognition, TTS, NLU, interactive dialogue management, and voice biometrics services are delivered through Nuance's hosted platform, Nuance On Demand.

The Nina Virtual Assistant SDK and cloud service is available in U.S., U.K., and Australian English, with additional languages scheduled to be made available later this year.


Page1 of 1