Ubitec Is A Good Example Of A NLU Agnostic Platform
Lately I have been considering how NLU-Engines are becoming less of a differentiator. Ubitec is a good case in point…
--
Intro To Ubitec
General
Ubitec is an Austrian based chatbot development framework with voice capabilities. Their focus is on-premise installation and language specific implementations.
I came across ubitec the first time while gleaning insights from the Gartner Peer Reviews Of Conversational AI Platforms. In the peer reviews, there are some fairly unknown frameworks including Ubitec, Laiyle Chatbot and others.
Implementations
Currently the Ubitec Bot Framework is doing most of its work for Government Organisations and the like; while Ubitec does virtually all of their work in the EMEA region.
Ubitec Bot Framework is also an outlier in the sense that they are focused on one area of Government implementations.
Access
In terms of platform access, Ubitec has very much the same approach followed by Boost AI, OneReach AI and a few other frameworks. Access to the framework is restricted and premised on a consultancy basis.
NLU-Agnostic
As seen below, within Ubitec a dynamic training pipeline can be defined to leverage various elements like translation, different NLU’s (in this case Snips NLU and Rasa).
Ubitec pride themselves on the fact that they have a standard standalone environment, but is easy configureable for interchangeable components and expansion.
There are five different methods of dialogue creation and management available in Ubitec. Ubitec has a unique approach to creating content, structuring dialogues and adding flexibility to conversations. One such feature is Ubitec’s contextualiser which manages the contextual awareness of the chatbot within the conversation.
Future Fragmentation
Dieter Perndl is very candid about the fact that Ubitec defaults to Snips NLU (open-source) and that he believes NLU per se is not where the opportunities are for optimisation. But rather in areas of design, orchestration and crafting customer specific solutions.
Dieter in his own words, “NLU itself is important but not the one crucial thing.
Training data is the more essential part making the NLU work properly.
At the end framework providers have to stay flexible with their components and architecture due to fast changing technology environments and the demand of customers being always on the latest stage of development and quality — especially for voicebots.”
Conclusion
Something I appreciated about Ubitec is that they do not exist within an a echo chamber. Ubitec is also not caught up in a parity-race with other frameworks…instead they are focussed on innovative ways to address current Conversation AI ailments.
I’m currently the Chief Evangelist @ HumanFirst. I explore and write about all things at the intersection of AI and language; ranging from LLMs, Chatbots, Voicebots, Development Frameworks, Data-Centric latent spaces and more.
