Developers

Moving forward from Watson Developer Conference with Project Intu

Share this post:

On November 9th, IBM Watson released an experimental offering – Project Intu, Watson’s very first embodied cognition platform. Thank you to all the forward-thinking developers, creators and technology-enthusiasts who at joined us at Watson’s inaugural Developer Conference and participated in the initial launch. For the users who weren’t able to make the conference but still have had a chance to try it out, we hope you’ve had a great experience so far. As the first exclusive group of users for Project Intu, we want to make sure you have the latest and greatest news, information, and updates on this emerging technology.

 

So what’s happened since the Developer Conference? For starters, hundreds of users have registered on Intu’s Gateway, new users are using over 10 different API’s on Bluemix, and there have been millions of API calls over the past month. One user created a Spanish-speaking advisor that can answer questions about the history of Perú, places to visit, and even cognitive computing. Project Intu allowed the easy coordination of NLC, Conversation, SST and TTS services on a Nao robot, with future plans to incorporate Visual Recognition! Don’t have a Nao robot, try a Raspberry Pi instead! Learn more about how to implement Intu on a Raspberry Pi here.

 

Aside from using multiple services in a more seamless way, Intu also provides pre-trained capabilities or “behaviors” that act out responses. Have you wanted your embodiment to respond to a certain name? Or maybe let you know when it’s raining outside? Intu provides an additional cognitive layer to Watson’s ability to understand, reason and respond – Watson can now behave, create a presence. Intu is the system that coordinates the cognitive mind (cognitive services) with the body (embodiment) through which it can express itself. Following an agent-based models, these behaviors are completed by individual agents working together. To learn more about behaviors and creating your own, watch this video.

Maybe it’s not just new behaviors you want to create. You can also create your own instance of Conversation to train Watson on new intents, entities, and dialog. Create a new bot that automates responses to your company’s FAQs, becomes a travel advisor in Perú, or even just tells you your favorite jokes – all deployed through the embodiment of your choice. All you have to do is change the workspace IDs in the Intu Gateway to direct to your newly trained bot.

And finally, we’ve been listening to the early feedback and addressed several of the initial bugs around connection and access issues. To further the improvements , please fill out our super quick 3-minute survey. We want to make sure we deliver excellent experiences, so we need all the feedback we can get

 

 

More Developers stories

GM Financial uses IBM Watson Assistant to develop a secure and powerful AI assistant

August 12, 2020 | AI for the Enterprise, Conversational Services

At GM Financial, it’s my job to drive remarkable customer experiences. Answering customers’ questions quickly and accurately is a big part of that. These days, much of our customer care comes in the form of live messages on our customer service app. People text us about things like the status of their remaining loan balances, ...read more


Uncover Hidden Risks with AI-Enabled GRC Platform

August 11, 2020 | AI for the Enterprise

Even before the added pressures caused by COVID-19, we had observed a considerable rise in regulatory fines and regulatory pressures across geographies and industries, and we expect this trend to continue. In financial services in 2019 alone, according to Fenergo research, $8.4 billion in fines were issued globally for non-compliance with AML, KYC and sanctions ...read more


Solving common challenges in sentiment analysis with help from Project Debater

August 10, 2020 | AI for the Enterprise

Today, we are going to talk about a feature that isn’t always talked a lot in the AI space: sentiment analysis. What is sentiment analysis? Sentiment analysis is a subset of natural language processing (NLP) capabilities that provides high level filters for users when exploring and evaluating data. Popularly, sentiment analysis is used to construct ...read more