As part of an ongoing effort to build a developer ecosystem around the IBM Watson platform for creating cognitive computing applications, IBM is now beta testing five additional Watson APIs.
Lauri Saft, director of the Watson ecosystem for IBM, says IBM has now defined eight APIs for Watson. The new API services provide access to functions such as speech-to-text; text to speech; visual recognition of various types of media content; conceptual search to identify explicit and implicit links between data; and tradeoff analytics that enable an application to balance conflicting goals against several sets of criteria.
The IBM Watson Developer Cloud running on the IBM Bluemix cloud platform, says Saft, has already spawned 6,000 application projects — 147 of which IBM has committed to support and sell. As a result, Saft notes that IBM is not only committed to helping developers build applications, it also is enabling developers to take advantage of multiple routes to market that IBM and its business partners can provide. In fact, IBM has created an entire business unit dedicated to the IBM Watson platform.
IBM is trying to pioneer not only new classes of applications, but also APIs that continuously learn as new data sources are added to the environment. Longer term, IBM is also working with partners such as Twitter to make it easier to aggregate data via APIs that can be analyzed using Watson applications. Put those capabilities together and it becomes apparent that all kinds of processes relating to research can be easily automated. Once the data is collected and analyzed, the information can then be used to automate any number of downstream business processes. While cognitive computing applications might eliminate many data collection tasks associated with entry-level jobs across a wide variety of professions, the ability to easily correlate data is expected to give rise to a broad range of higher level application services that should serve to make professionals in those fields much more productive.
While IBM has created a lot of momentum around a platform that combines natural language and machine learning to create cognitive computing applications that can consume a massive amount of data, rivals such as Microsoft along with a host of startups are now pouring massive amounts of research and development dollars into this category. As a result, it’s only a matter of time before a raft of cognitive computing applications come to market that not only change the way data is consumed, but like every application that has gone before, they will also need to be integrated with one another.