Creating a Release

Creating a release enables your bot to interact with users through the chosen platforms – such as Phone or Textchat. This process involves deploying your bot's configuration, ensuring users can engage with it via calls or chat sessions. Additionally, the release now includes the option to apply global Prosody settings, enhancing voice consistency across all interactions.

Phone Platform Deployment

The following outlines how to deploy a project on the Phone platform.This is

Step 1 – Create a New Release

Navigate to Deployments -> select Releases, and click Add New Release:

The following displays:

New Release – Available Platforms

Step 2 – Select the Platform

For a phone bot, pick 'Phone 2' from the list of Available Platforms, our most current phone infrastructure. For chat bots, choose 'Chat 2' for the latest chat features.

1. Once the appropriate platform is selected, click Next to proceed

Older platforms like 'Phone 1' and 'Chat 1' are outdated and no longer supported.

Step 3 – Configure Platform Settings

1. Enter the following mandatory details:

   • SIP Username

   • SIP Password

   • SIP Registrar Hostname

   

   New Release – Platform Settings

The Caller Language dropdown menu shows the language you've previously set.

1. Click Next.

Step 4 – Configure Release Settings

At this stage, you can configure various optional and mandatory settings:

New Release – Release Settings

The following list corresponds to the numbered image above:

1 – Select a Dialog Version

A release is intrinsically linked to a specific version as it determines the flow that will be accessible on a platform. By default, a release selects the current Draft Version of the project. However, you should verify that the correct version is actively chosen to ensure the intended flow is deployed.

2 – Choose an Environment

Selecting an environment is optional, but as a best practice, it's advisable to always specify the appropriate environment that corresponds with the stage of your project's lifecycle.

3 – Select an Intent Classification

If your contract includes the use of either Description or Utterances for intent classification, ensure to select the method provided. If this feature is not yet available to you, you may skip this step.

4 – Select Caller Language

It is mandatory to select the caller's language to enable the Natural Language Understanding (NLU) system to correctly process user input.

5 – Select Speaker Voice

For phone deployments, choosing the bot’s language and accent is mandatory. This setting is essential for the TTS engine to generate the correct voice interactions.

For chat deployments, this step is not required as the language selection is unified.

6 – Global Prosody Settings

While this step remains optional, toggling ON global Prosody settings allows you to fine-tune the pitch, rate, and volume of the bot's voice. Refer to the Prosody SSML tag documentation for detailed guidance.

7 – Release Description

As a recommended practice, include the environment (Dev/Stage/Prod) and the method of contact (phone number or chat link) within the description to document clearly how the bot can be accessed.

8 – Toggle Production Release

This toggle is mandatory only when preparing for a production release. Switching it ON ensures the bot is allocated the necessary resources to perform optimally in a live environment.

9 – Proceed to Analytics

After ensuring all mandatory fields are correctly filled, click Next to move to the Analytics tab, which is crucial for monitoring the performance of your bot.

Step 5 – Analytics Configuration

Modify any analytics-related fields to match your needs and finalize by clicking Create.

New Release – Analytics

Textchat Platform Deployment

To deploy a project in the Textchat platform:

Step 1 – Initiate a New Release

Follow the same first step as the Phone platform.

Step 2 – Select Platform

Choose the Textchat 2 from the Available Platforms list and click Next.

New Release – Available Platforms – Textchat

Step 3 – Secure Your Release

Optionally, enhance the security of your release endpoint by activating the Protect my release endpoint with an API Key option. An API key will be generated, not only safeguarding your release endpoint against unauthorized third-party access but also enabling the connection of this bot to a widget. Click Next to proceed.

New Release – Platform Settings – Textchat

Remember to secure your API key and not share it publicly to maintain the integrity of your release.

Step 4 – Define Release Settings

New Release – Release Settings – Textchat

The following list corresponds to the numbered image above:

Number	Description
1	Choose the version for your release from the dropdown menu, such as Version 01 - Draft.
2	Specify the environment that this release is targeting by selecting from the Environment dropdown.
3	Confirm the language of your release, for instance, English (en-US) should be pre-selected if it's your primary language.
4	Provide a concise description of your release in the Description field for reference.
5	If this release is intended for real customer interactions, switch the Production Release toggle ON to ensure it receives the necessary resources and monitoring for reliable availability.

After configuring these settings, click Next.

Step 5 – Enable Analytics and Training Data

• Enable Training Data – Toggle this option ON to record interactions with your release, which is crucial for tracking and analyzing user engagement. For phone deployments, enabling this setting allows the system to capture user utterances

• Integrate Analytics Services – To gain insights from user interactions, toggle ON the option to send requests to your Analytics Services. This integration is crucial for all platforms, including Phone.

After setting up these options, click Create to finalize the release process.

A release doesn't equate to a public launch. It's a developer's version of your project for testing purposes on the selected platform.