Maluuba is a company that has created a patent-pending, voice recognition and language engine. According to Crunchbase, Samsung Ventures funded $2 million to Maluuba in February 2012, and the Maluuba language engine "is a product of two years of advanced research in Artificial Intelligence, Machine Learning and linguistics." Yesterday, Maluuba announced the introduction of a preview alpha release of what the company describes as the "first of its kind" natural language understanding API (nAPI).
The Natural Language Understanding API makes it possible for developers to create applications that include language recognition functionality.
The ability to complete web forms via speech instead of manually typing , or simply asking a question verbally instead of performing a manual search, are only two examples of how voice recognition technology can benefit application users.
One example of how nAPI could be used in applications was provided in the announcement:
"As an example, a user could say “Get me a table for 2 at an Italian restaurant in San Francisco” and nAPI will do the hard work of understanding and parsing out all of the relevant parts of their search. You’ll be provided with the intent “reserve table”, the number of people for the reservation as “2”, the type of restaurant as “Italian”, and the location as “San Francisco”."
Currently (nAPI) consists of two main components; the Interpret API which interprets the meaning of phrases and the Normalize API which finds normalized values for dates and times. Simple wrappers for the Maluuba nAPI are available on GitHub and the languages available at this time are Ruby, Java and Python.
The nAPI supports 22 different feature domains including business, entertainment, music, calendar, travel, weather and search. A complete categories list is provided on the Natural Language API Developers Site.
Joshua Pantony, CTO at Maluuba says about nAPI in the announcement that:
"This API is the beginning of something great, something that is going to completely change how people interact with their devices. But it is just a beginning. We are starting out with the given domains, but plan on massively expanding it as feature requests come in, and we see how people want to use this technology."
Developers interested in using nAPI in their applications must submit a request and receive pre-approval before accessing the API.