Cognigy.AI 3.6 Released

Topic: 
|
4 min read
Philipp Heltewig
Media contact: Philipp Heltewig April 8, 2020

Düsseldorf (GER), March 30, 2020 +++ Today we are announcing the release of Cognigy.AI 3.6 - the newest version of our industry-leading Conversational Automation Platform.

Cognigy.AI 3.6 is the last release before we release Cognigy.AI 4.0 in a few months, but still jam-packed with amazing new features:

  • Finnish & Swedish NLU
  • NLU Transformers
  • Extended Endpoint Transformers
  • Custom Process Elements
  • Profile & Context TTL for Compliance

Here are the details…

Finnish & Swedish NLU

Cognigy.NLU is known as one of the leading Natural Language Understanding engines in the industry and we're always thriving to deliver the best possible technology to our many customers around the world. With this release, we're adding native NLU models for Finnish & Swedish to Cognigy.AI, enabling the creation of advanced Conversational AI in these languages and bringing the total number of supported languages to 22.

NLU Transformers

With Cognigy.AI 3.5 we released the powerful and novel concept of Endpoint Transformers, which let you hook, via code, into the processing of messages that come from and go back out to a channel.

With Cognigy.AI 3.6 we're extending this concept to NLU Connectors, essentially allowing developers to influence the NLU pipeline through JavaScript/TypeScript code or to connect to any third party NLU of their choice.

Using this feature, developers could, for example, easily connect to RASA NLU, Salesforce Einstein Language or any other NLU, whilst still leveraging the full power of the Cognigy.AI Flow Editor and other graphical tools.

Developers could also configure Cognigy.AI to run Cognigy.NLU, but augment the results with third party sentiment analysis or other advanced machine learning services in order to create even more outstanding conversational automation solutions.

Extended Endpoint Transformers

Cognigy.AI's Endpoint Transformer feature has been extended to now also include non-POST verbs, meaning that all requests to a REST and Webhook endpoint (including GET, PUT, etc) can be intercepted. This can be useful when replying to verification requests from external APIs.

Custom Process Elements

Cognigy.AI Processes allow the easy creation of linear, stepped conversations. Until this release, the available question types were Text, YesNo, DateTime, Number, Email,  Phone Number and Keyphrase. With this release, we're adding a "Custom" question type, in which the editor can define the question, validation and also trigger the UI on the front end. This way, custom webchat plugins can be used from within Processes.

Profile & Context TTL for Compliance

Cognigy is offering a wide variety of privacy-related tools, from Endpoint Data Management tools, over fine-granular in-session privacy controls (Blind Mode) to Contact-specific privacy tools. In this release, we're adding the ability for system-admins to define TTL (time-to-live) thresholds after which the persistent Context and Profile data stores will be flushed. This way, Cognigy.AI can be configured to, for example, remove all stale Context from the database after 24 hours and all unused Contact Profiles after a period of 30 days.

Summary

Our solution makes it easier than ever to build, deploy and manage Conversational AIs at scale from a centralized conversational platform. Our tools enable business users to develop solutions in customer experience, customer support and process automation, while giving developers the power to customize the solution to their liking. Cognigy.AI 3.6 adds further to this vision and we’re looking forward to seeing what our customers and partners around the world will build with it. 

Start your free trial
image119-1
image119-1
image119-1