> ## Documentation Index
> Fetch the complete documentation index at: https://docs.cognigy.com/llms.txt
> Use this file to discover all available pages before exploring further.

# 4.91

<Warning>
  With this release, the Understood Messages concept and its logic in the [NLU Performance](/insights/reports/nlu-performance) and [Intent Trainer](/ai/platform-features/nlu/intents/intent-trainer) will be deprecated. Use the [Goals and Tasks](/ai/agents/analyze/goals-and-tasks/overview) concepts to evaluate whether your AI Agent understood a particular phrase.

  Refer to the list of all upcoming removals on the [Deprecations and Removals](/release-notes/deprecations-and-removals) page.
</Warning>

## 4.91.0

### Cognigy.AI

**Released** December 10, 2024

#### Features

* [Agentic AI Beta](/ai/agents/overview) is publicly available for all users. To manage a new generation of your AI Agent Workforce, Cognigy.AI introduces the [AI Agent Management](/ai/agents/develop/manage-ai-agents) interface. Agentic AI represents a new generation of autonomous, goal-oriented AI Agents designed to boost operational efficiency, enhance customer satisfaction, and drive superior enterprise outcomes
* Added the capability to [enforce login with Single Sign-on (SSO)](/ai/administer/installation/single-sign-on-saml2#enforce-login-with-sso), enhancing security and simplifying access management

#### Improvements

* Added [on-the-fly resizing for uploaded images](/ai/agents/deploy/endpoints/file-storage#resize-images)
* Updated the [Project Creation](/ai/agents/develop/projects-and-flows/projects) page. You can now select the type of project to create: Agentic AI or Intent-driven AI
* Removed the restriction on the maximum number of buttons for the Text with Buttons output type in Say Nodes

#### Bug Fixes

* Fixed the issue related to the Triton LABS encoder by updating the embedding package
* Fixed the issue where execution stops after encountering `null` from the output transformer, and 0 was not being sent as the text to the user in Webchat v2 and v3
* Fixed the issue where handoverEscalations analytics were not updated correctly
* Fixed the issue where disabling the Enable Call Event: Call in Progress generic setting did not hide the section for configuring the handler
* Fixed the issue where the Blind Mode Node was incorrectly applied during the processing of the previous message
* Fixed the issue where the LLM provided by Google Gemini returned the entire response at once, even when it was configured to stream the output
* Fixed the issue of overlapping records in the Intent Trainer

### Cognigy Voice Gateway

**Released** December 10, 2024

#### Improvements

* Added the capability of setting User Input Timeout, Flow Input Timeout, Recognizer, Synthesizer, DTMF, Barge-In, and Continuous ASR with CognigyScript in the [additional session parameters](/ai/agents/develop/node-reference/voice/voice-gateway/parameter-details)
* Added the `da-DK-Standard-F` and `da-DK-Standard-G` voices when using [Google Speech Services for Text to Speech](/ai/agents/test/voice-preview#set-up-a-speech-provider)
* Added the capability to skip fallback speech providers with speech provider failover by setting the `skipFallbackSpeechProviders=true` flag during configuration. You can enable the speech provider failover in the [Voice Gateway Endpoint Settings](/ai/agents/deploy/endpoint-reference/voice-gateway)

#### Bug Fixes

* Fixed the issue where fallback transfer calls by dial or refer would not hang up if the call failed
* Fixed the issue where disabling the `Enable Call Event: Call in Progress` setting did not hide the `Call Event Action: Call in Progress` section for configuring the handler
* Fixed the issue where application data was missing when creating an outbound call. This issue led to the wrong selection of speech credentials
* Fixed the issue where the private SIP address was not available in the `from` header when private networking was enabled for that CIDR
* Fixed the issue where the DNS resolving for private networking was not working

### Webchat

**Released** December 10, 2024

#### Improvements

* Improved the UI of Webchat v3 by displaying uploaded images as previews in the chat
* Updated Webchat v3 and its Demo Webchat to [v3.8.1](https://github.com/Cognigy/Webchat/releases/tag/v3.8.1)

### xApps

**Released** December 10, 2024

#### Improvements

* Improved the content structure of messages sent from xApps in the conversation chat within Cognigy Live Agent to enhance readability for human agents

### Cognigy Live Agent

**Released** December 10, 2024

#### Bug Fixes

* Fixed the issue with invalid Project IDs in the [Inbox Settings](/live-agent/settings/inboxes). Previously, you could enter an invalid Project ID in the Inbox Settings and no errors would be raised. Now, a validation step does not allow entering an invalid Project ID
* Fixed the issue where the validation of custom sound URLs was not working correctly

### Cognigy Insights

**Released** December 10, 2024

#### Bug Fixes

* Fixed the issue where the Gallery output type is not displayed in the Transcript Explorer

### Infrastructure

#### Cognigy.AI: Runtime and IDE File Shares

To deploy the Agentic AI feature, file shares named `runtime` and `ide` are required. The procedure for deploying file shares on Amazon Elastic Kubernetes Service (EKS) and Azure Kubernetes Service (AKS) is explained below.

##### Amazon Elastic Kubernetes Service (EKS)

1. Create `runtime` and `ide` Amazon Elastic File System (EFS).
2. Add the following code to the Cognigy.AI `values.yaml` file:

   ```yaml theme={null}
   storageClass:
     aws:
       efs:
         ide:
           enabled: true
           fileSystemId: "<ide_efs_id>"
         runtime:
           enabled: true
           fileSystemId: "<runtime_efs_id>"

   ## PVC for new nfs
   ##
   persistentVolumeClaim:
     enabled: true
     runtime:
       enabled: true
     ide:
       enabled: true
   ```

##### Azure Kubernetes Service (AKS)

Add the following code to the Cognigy.AI `values.yaml` file:

```yaml theme={null}
storageClass:
  azure:
    file:
      runtime:
        enabled: true
      ide:
        enabled: true

## PVC for new nfs
##
persistentVolumeClaim:
  enabled: true
  runtime:
    enabled: true
    storage: 1000Gi
  ide:
    enabled: true
    storage: 1000Gi
```

##### Other Cloud Providers

The Cognigy.AI Helm chart doesn't provide out-of-the-box support for other cloud providers or on-premises environments. To configure `runtime` and `ide` file shares for environments other than EKS and AKS, follow these steps:

1. Create `storageClass` for both `runtime` and `ide`.
2. Enable `runtime` and `ide` in the `Values.persistentVolumeClaim` section. Specify `storageClassName` created in the previous step as follows:
   * `Values.persistentVolumeClaim.runtime.storageClassName`
   * `Values.persistentVolumeClaim.ide.storageClassName`

#### Cognigy Voice Gateway

* Updated `vg-drachtio-server` to `0.8.27-rc17`

#### Version Compatibility Matrix

The version compatibility matrix was updated for the following Cognigy products:

* [Cognigy.AI, Insights, and Agent Copilot](/ai/administer/installation/version-compatibility-matrix)
* [Cognigy Live Agent](/live-agent/installation/deployment/version-compatibility-matrix)
* [Cognigy Voice Gateway](/voice-gateway/installation/version-compatibility-matrix)