KB Articles – AI Voice Portal

1. Creating an AI Voice Agent

Overview

This guide explains how to create a new AI Voice Agent in the portal. Agents can be created directly from the Dashboard or through the Call Flows section.

Steps to Create an Agent

Option 1 – From the Dashboard

1. From the Dashboard, click Create Agent.
2. The Agent Editor will open — configure the agent as required.

Option 2 – From Call Flows

1. Navigate to Call Flows in the left-hand menu.
2. Click New Call Flow, then select From Scratch.
3. The Agent Editor will open — configure the agent as required.

Naming and Saving

4. Enter a name for the agent in the name field at the top of the editor.
5. Make all required edits and changes.
6. Click Save.

Verify the Agent

7. Return to Call Flows.
8. Confirm the new agent appears in the list.

Key Notes

🎥 Watch the demo video below

2. Duplicating an Agent

Overview

This guide explains how to duplicate an existing AI Voice Agent (Call Flow) in the portal. Duplicating an agent creates an independent copy that can be edited without affecting the original.

Steps to Duplicate an Agent

1. Navigate to Call Flows in the left-hand menu.
2. Click the three dots (⋮) next to the agent you want to duplicate.
3. Select Duplicate.
4. Confirm that the destination workspace is the same workspace you are currently in.
5. The duplicated agent will appear in the Call Flows list.
6. Open the duplicated agent and make any required edits.

Key Notes

🎥 Watch the demo video below

3. Post Call Analysis

Overview

This guide explains how to configure Post Call Analysis for an AI Voice Agent. Post Call Analysis allows you to define specific data points to be extracted automatically from each call, which are then available in call history, webhook payloads, post call emails, and post call SMS messages.

Steps to Configure Post Call Analysis

1. Navigate to Call Flows and locate the agent you want to edit.
2. Ensure you are in the Advanced View.
3. Go to Data and click Post Call Analysis.
4. Click Add to create a new data point.
5. Enter a title for the variable (e.g., Name, Email Address).
6. In the description field, write a brief instruction for what the AI should extract — treat this as a mini-prompt (e.g., The caller’s first and last name as provided during the call).
7. Repeat for each data point you want to capture.
8. Optionally, change the Analysis Model to select a different LLM for extraction.
9. Click Save when finished.

Key Notes

🎥 Watch the demo video below

4. Background Sounds & Hold Music

Overview

This guide explains how to configure background audio for an AI Voice Agent, including hold music, thinking sounds, and ambient audio. You can use the platform’s built-in audio options or upload your own.

Steps to Configure Background Audio

1. Navigate to Call Flows and locate the agent you want to edit.
2. Ensure you are in the Advanced View.
3. Go to Voice and click Background Audio.
4. From here you can configure:
   • Hold Music – audio played while the caller is on hold
   • Thinking Sound – audio played while the agent is processing a response
   • Ambient Sound – background audio present throughout the call
5. To use custom audio, click Manage and upload your own audio file.
6. Adjust the volume for each audio type using the volume controls.
7. Hover over the information icon (?) next to any setting for additional detail.
8. Once all changes are made, click Save to Draft or Publish to Production.

Key Notes

🎥 Watch the demo video below

(Insert video embed here)

5. Editing Office Hours

Overview

This guide explains how to configure the office hours for an AI Voice Agent. Office hours control when the agent is active and able to make or receive calls. You can set active days and times, configure the time zone, and block out specific dates such as public holidays.

Steps to Configure Office Hours

1. Navigate to Call Flows and locate the agent you want to edit.
2. Ensure you are in the Advanced View.
3. Go to Call Handling and click Office Hours.
4. Toggle Office Hours on to enable scheduling restrictions.
5. Set the active days by toggling each day on or off.
6. Set the active time range for each day.
7. Select the correct time zone for the agent.
8. To block specific dates (e.g., public holidays), click the Custom Dates option and select the relevant dates.
9. Once all changes are made, click Save to Draft or Publish to Production.

Key Notes

🎥 Watch the demo video below

6. Agent Dictionary

Overview

This guide explains how to use the Agent Dictionary to improve transcription accuracy for your AI Voice Agent. The dictionary allows you to add industry-specific words, proper nouns, and acronyms that the agent may otherwise mishear or transcribe incorrectly.

Steps to Configure the Agent Dictionary

1. Navigate to Call Flows and locate the agent you want to edit.
2. Ensure you are in the Advanced View.
3. Go to Voice and click Agent Dictionary.
4. Type in any words or acronyms you want to add to the dictionary.
5. Once all words have been entered, click Save to Draft or Publish to Production.
6. Hover over the information icon (?) for additional guidance on the dictionary feature.

Key Notes

🎥 Watch the demo video below

7. Editing Call Timings & Reminders

Overview

This guide explains how to configure call timing settings and silence reminders for an AI Voice Agent. These settings control how the agent handles periods of silence, how long calls can run, and what the agent says when a caller becomes unresponsive.

Steps to Configure Call Timings

1. Navigate to Call Flows and locate the agent you want to edit.
2. Ensure you are in the Advanced View.
3. Go to Call Handling and click Call Timing.

Silence Timeout

4. Set the Silence Timeout duration — this is how many seconds of silence must pass before the agent speaks a reminder (e.g., 10 seconds).
5. Set the End Call After Timeout duration — this is how long the agent waits after delivering the reminder before ending the call.
6. Set the Reminder Retries — the maximum number of times the agent will repeat the reminder before ending the call (maximum of 3).
7. Click + to add one or more reminder messages. If multiple messages are added, the agent will cycle through them in order.

Maximum Call Duration

8. Set the Maximum Call Duration to define the longest a call can run.
9. Enter an end-of-call message that the agent will deliver when the maximum duration is reached.
10. Hover over the information icon (?) on any field for additional detail.
11. Once all changes are made, click Save to Draft or Publish to Production.

Key Notes

🎥 Watch the demo video below

8. Sending Booking Links via SMS

Overview

This guide explains how to configure an AI Voice Agent to send a booking link via SMS during or after a call. The booking link is typically sourced from a third-party scheduling tool and can be triggered by the agent when relevant during a conversation.

Steps to Configure a Booking Link

1. Navigate to Call Flows and locate the agent you want to edit.
2. Ensure you are in the Advanced View.
3. Go to Data and click SMS Actions.
4. Hover over the information icon (?) for additional detail on how this feature works.
5. Toggle the booking link option on.
6. Paste the booking link URL into the provided field (e.g., a Calendly link or similar scheduling tool).
7. Click Save.
8. Once all changes are made, click Save to Draft or Publish to Production.

Key Notes

🎥 Watch the demo video below

9. Linking a Meta Account

Overview

This guide explains how to connect a Meta (Facebook/Instagram) account to an AI Voice Agent using the Meta Lead Callback integration. This allows the agent to automatically call new leads the moment they submit a form via a Meta ad, significantly improving conversion rates.

Steps to Link a Meta Account

1. Navigate to Call Flows and locate the agent you want to connect.
2. Ensure you are in the Advanced View.
3. Go to Integrations.
4. Locate Meta Lead Callback and click Connect.
5. Follow the prompts to authenticate and connect your Meta account.
6. Once connected, click Close.
7. Click Save to Draft or Publish to Production.

Key Notes

🎥 Watch the demo video below

10. Post Call SMS & Email

Overview

This guide explains how to configure an AI Voice Agent to automatically send SMS messages and emails at the end of a call. Notifications can be sent to the business, the caller, or up to three email recipients.

Steps to Configure Post Call Notifications

1. Navigate to Call Flows and locate the agent you want to edit.
2. Ensure you are in the Advanced View.
3. Go to Data and click Post Call Actions.
4. Toggle on the notification types you want to enable:
   • SMS to Business – sends a summary SMS to the business owner or nominated number
   • SMS to Caller – sends a summary SMS to the person who was on the call
   • Email Notification – sends a summary email to up to three email addresses
5. Enter the relevant contact details for each enabled notification.
6. Once all changes are made, click Save to Draft or Publish to Production.

Key Notes

🎥 Watch the demo video below

11. Transfers

Overview

This guide explains how to configure call transfers for an AI Voice Agent. There are three transfer types available — Cold, Warm, and Whisper — each with different behaviour for how the receiving party is connected and informed.

Transfer Types

Steps to Configure a Transfer

1. Navigate to Call Flows and locate the agent you want to edit.
2. Ensure you are in the Advanced View.
3. Go to Call Handling and locate the Transfers section.
4. Click Add Transfer and give it a name (e.g., Support).
5. Select the transfer type — Cold, Warm, or Whisper.
6. Enter the destination number the call should be transferred to.
7. For Warm and Whisper transfers, click Advanced Settings and enter the message the agent will deliver to the receiving party before connecting (e.g., caller’s name, company, reason for calling).
8. Set a timeout duration — if the receiving party does not answer within this time, the agent will return to the caller.
9. Enter the timeout message the agent will deliver if no one answers (e.g., Unfortunately no one is available right now — let me continue to assist you).
10. Select whether the agent should continue the call or end the call if the transfer times out.
11. Configure the transfer conditions in the call flow prompt to define when the agent should invoke the transfer.
12. Click Save to Draft or Publish to Production.

Key Notes

🎥 Watch the demo video below

12. Toggle Prior Calls Context

Overview

This guide explains how to enable or disable Prior Calls Context (AI Memory) for an AI Voice Agent. When enabled, the agent can reference information from previous calls made from the same phone number, allowing for more personalised and context-aware conversations.

Steps to Toggle Prior Calls Context

1. Navigate to Call Flows and locate the agent you want to edit.
2. Ensure you are in the Advanced View.
3. Go to General and click AI Memory.
4. Hover over the information icons (?) to review how each setting works.
5. Toggle Prior Calls Context on or off as required.
6. Click Save to Draft or Publish to Production.

Key Notes

🎥 Watch the demo video below

13. Sharing a Demo Link

Overview

This guide explains how to generate and share a demo link for an AI Voice Agent. Demo links allow anyone to interact with an agent via a web browser without needing a phone number or phone call, making them ideal for demonstrations and testing.

Steps to Create and Share a Demo Link

1. Navigate to Call Flows and locate the agent you want to share.
2. Click Share Demo at the top of the agent editor.
3. Enter a name for the demo link.
4. Expand Advanced Settings to configure optional limits:
   • Expiry Date – select a preset duration or enter a custom date for when the link will expire
   • Per User Limit – the maximum call duration allowed per individual user session
   • Total Budget – the combined total minutes available across all users of this link
5. Click Create to generate the link.
6. Click Confirm and Save.
7. Return to the Demo section, copy the link, and share it as required.

Key Notes

🎥 Watch the demo video below

14. Changing the LLM (AI Model)

Overview

This guide explains how to change the Large Language Model (LLM) used by an AI Voice Agent. The LLM is the core AI model that powers the agent’s understanding and responses. Different models offer different capabilities, and the choice of model also affects which voices are available.

Steps to Change the LLM

1. Navigate to Call Flows and locate the agent you want to edit.
2. Ensure you are in the Advanced View.
3. Locate the LLM selector within the agent settings.
4. Choose from the available models:
   • OpenAI models (e.g., GPT series)
   • Google models (e.g., Gemini)
   • Open source models
   • Speech-to-speech models
5. Once selected, click Save to Draft or Publish to Production.

Voice Compatibility by Model Type

Key Notes

🎥 Watch the demo video below

15. Viewing Analytics

Overview

This guide explains how to access and interpret the analytics dashboard in the portal. Analytics provide visibility into call volumes, agent activity, and call outcomes across your workspace.

Steps to View Analytics

1. Navigate to Analytics in the left-hand sidebar.
2. The analytics dashboard will display key metrics including:
   • Call volumes by time of day
   • Calls per agent (inbound and outbound)
3. Use the agent filter to view data for a single agent or across all agents.
4. Use the date range selector to adjust the reporting period.
5. Add dispositions to categorise and track call outcomes (e.g., Completed, Transferred, Voicemail).
6. Assign each agent a nickname for easier identification within the dashboard.

Key Notes

🎥 Watch the demo video below

16. How to Test an Agent

Overview

This guide explains how to test an AI Voice Agent directly within the portal using either voice or text. Testing allows you to interact with the agent before it goes live and identify any issues with the prompt or configuration.

Steps to Test an Agent

1. Navigate to Call Flows and locate the agent you want to test.
2. Click Voice to initiate a voice conversation with the agent.
3. The agent will connect and you can speak with it as a caller would.
4. Alternatively, use the text chat option to type messages and receive responses in text format.
5. Review how the agent responds and make note of any issues.
6. Return to the agent settings or prompt to make adjustments as needed.

Key Notes

🎥 Watch the demo video below

17. Dynamic Variables whilst Testing

Overview

This guide explains how to set and manage dynamic variables when testing an AI Voice Agent in the portal. Dynamic variables are values injected into the agent’s prompt at runtime — for example, a caller’s first name — and must be populated before testing to ensure the agent behaves correctly.

What Are Dynamic Variables?

Dynamic variables appear in the agent prompt as orange highlighted text enclosed in curly brackets (e.g., {first_name}). They must be:

• A single word (no spaces)
• Case sensitive
• Defined in the Variables panel before testing

Steps to Set Dynamic Variables for Testing

1. Navigate to Call Flows and locate the agent you want to test.
2. Confirm the agent prompt contains dynamic variables (displayed in orange with curly brackets).
3. Before testing, go to the Variables panel.
4. Locate the variable (e.g., {first_name}) and enter a test value (e.g., John).
5. Optionally, click Save to Library to store this variable set for future test sessions — enter a name and confirm.
6. In the Library, click the star icon next to a saved set to make it the default — this means the agent will automatically load that variable set each time you test.
7. Click Load to apply the variable set before testing.
8. Only add variables that exist in the agent prompt — adding variables not referenced in the prompt has no effect.
9. Once testing is complete, save any changes using Save to Draft.

Key Notes

🎥 Watch the demo video below

18. Changing the Agent Voice

Overview

This guide explains how to change the voice used by an AI Voice Agent. The portal includes a range of built-in voices as well as ElevenLabs voices. You can filter by characteristics such as accent, gender, or age, and preview voices before selecting.

Steps to Change the Agent Voice

1. Navigate to Call Flows and locate the agent you want to edit.
2. Ensure you are in the Advanced View.
3. Go to Voice and click on the Voice selector.
4. Browse the available voices or use the search field to filter by characteristics (e.g., Australian, Female, Young).
5. Click Play next to any voice to preview it.
6. To test a voice in a full conversation, click Voice and speak with the agent directly.
7. Select the voice you want to use and click Close.
8. Click Save to Draft or Publish to Production.

Voice Types Available

Key Notes

🎥 Watch the demo video below

19. Changing the Agent Language

Overview

This guide explains how to change the language used by an AI Voice Agent. The platform supports multiple languages and, when using speech-to-speech models, can switch between languages dynamically during a call.

Steps to Change the Agent Language

1. Navigate to Call Flows and locate the agent you want to edit.
2. Ensure you are in the Advanced View.
3. Locate the Language selector and choose the required language.
4. Ensure the voice assigned to the agent matches the selected language — using an English-accent voice for a non-English language will produce unnatural results.
5. Where possible, write the agent’s prompt in the same language as the selected language setting for the most natural output.
6. Click Save to Draft or Publish to Production.

Key Notes

🎥 Watch the demo video below

20. Editing Voice Speed

Overview

This guide explains how to adjust the speaking speed of an AI Voice Agent. Voice speed can be increased or decreased to better suit the agent’s use case and target audience.

Steps to Edit Voice Speed

1. Navigate to Call Flows and locate the agent you want to edit.
2. Ensure you are in the Advanced View.
3. Go to Voice and locate the Voice Speed control.
4. Adjust the speed up or down to the desired level.
5. Click Save to Draft or Publish to Production.
6. Use the Voice test button to speak with the agent and confirm the speed feels natural.

Key Notes

🎥 Watch the demo video below

KB Articles – AI Voice Portal (Batch 2)

21. Editing the System Prompt

Overview

This guide explains how to access and edit the main system prompt for an AI Voice Agent. The system prompt is the core instruction set that controls how the agent behaves, what it says, and when it executes actions such as transfers or API calls.

Steps to Edit the System Prompt

1. Navigate to Call Flows and locate the agent you want to edit.
2. Click into the agent to open the editor.
3. The system prompt is displayed in the main editor panel.
4. To expand the editor to full screen, click the expand icon.
5. Make any required edits to the prompt.
6. To preview the prompt structure visually, switch to the Visualiser view — this shows how each section of the call links together and when different parts of the prompt are triggered.
7. Once edits are complete, click Save to Draft to test before going live, or Publish to Production to apply immediately.

Key Notes

🎥 Watch the demo video below

22. Changing Agent Name & Call Direction

Overview

This guide explains how to rename an AI Voice Agent and change the direction in which it can handle calls — inbound, outbound, or both.

Steps to Change Agent Name and Direction

1. Navigate to Call Flows and locate the agent you want to edit.
2. Ensure you are in the Advanced View.
3. Click the name field at the top of the editor to rename the Call Flow.
4. To assign the agent a separate display name or persona name (e.g., Alex, Samantha), enter it in the Agent Nickname field.
5. Locate the Call Direction setting and select the appropriate option:
   • Outbound only – the agent can only make outbound calls
   • Inbound and Outbound – the agent can both make and receive calls
6. Click Save to Draft or Publish to Production.

Key Notes

🎥 Watch the demo video below

23. Configuring In-Call API Calls

Overview

This guide explains how to configure in-call API calls for an AI Voice Agent. API calls allow the agent to retrieve or send data to external systems in real time during a conversation — for example, checking product availability, looking up a customer record, or submitting a form.

Steps to Configure an In-Call API

1. Navigate to Call Flows and locate the agent you want to edit.
2. Ensure you are in the Advanced View.
3. Locate the API section within the agent settings.
4. Click Add API and configure the following fields:
   • Method – the HTTP method (e.g., GET, POST)
   • Endpoint – the URL the API call will be sent to
   • Name – a label used to reference this API call within the prompt
   • Variables – any dynamic variables to be passed with the request
   • Authentication – add authentication credentials if required
   • Request Headers – add any required headers
5. Optionally, configure a Text to Speak message that the agent will say while the API call is being processed (e.g., “Just give me a moment to check that for you”) — this prevents silence while the system waits for a response.
6. Configure whether the API should run pre-call (before the conversation starts) or be triggered during the call based on prompt conditions.
7. Reference the API by its Name within the system prompt to define when it should be invoked.
8. Click Save to Draft or Publish to Production.

Key Notes

🎥 Watch the demo video below

24. Configuring Webhooks

Overview

This guide explains how to configure system webhooks for an AI Voice Agent. Webhooks allow the platform to send call data to external systems automatically — either before a call begins, after a call ends, or when the call status changes.

Webhook Types

Steps to Configure a Webhook

1. Navigate to Call Flows and locate the agent you want to edit.
2. Go to Data and click System Webhooks.
3. Select the webhook type you want to configure.
4. Enter the destination URL where the webhook payload should be sent.
5. For the post-call webhook, customise the payload to include the specific data fields you need.
6. Hover over the information icon (?) on each field for additional guidance.
7. Click Save to Draft or Publish to Production.

Key Notes

🎥 Watch the demo video below

25. Recording & Security Settings

Overview

This guide explains how to configure the security and compliance settings for an AI Voice Agent, including call recording, compliance monitoring, anti-bot detection, HIPAA mode, data retention, and transcription redaction.

Steps to Access Security Settings

1. Navigate to Call Flows and locate the agent you want to edit.
2. Ensure you are in the Advanced View.
3. Click on the Security tab.

Available Security Settings

Recording

4. Toggle Recording on to enable call recording for all calls and web demo sessions.
5. Toggle Public Recording Link on if you want a publicly accessible URL to be generated for each recording — this is required if you intend to pass recording links to external systems such as a CRM.

Compliance Monitoring

6. Toggle Compliance on to enable compliance monitoring across inbound and outbound calls. Compliance results are visible in the call history.

Anti-Bot Detection

7. Toggle Anti-Bot Detection on to detect whether an AI-generated voice is calling in. Hover over the information icon for further detail.

HIPAA Mode

8. Toggle HIPAA on to enable HIPAA-compliant data handling. When enabled:
   • All call data (transcripts, recordings, LLM processing history) is wiped from the platform and its underlying services immediately after the call ends and the post-call webhook fires
   • Data can still be sent to an external CRM via webhook before it is wiped
   • Call history and transcripts will not be accessible in the portal after the call
   • Prior Calls Context (AI Memory) cannot be used, as there is no retained call data to reference

Transcription Redaction

9. Toggle Transcription Redaction on and configure which data points should be redacted from transcripts (e.g., date of birth, caller name).

Data Retention

10. Configure the Data Retention period for transcripts, summaries, and recordings — options range from 1 day to 1 year, after which data is automatically deleted.
11. Click Save to Draft or Publish to Production.

Key Notes

🎥 Watch the demo video below

26. Connecting GoHighLevel

Overview

This guide explains how to connect a GoHighLevel (GHL) account to an AI Voice Agent. Once connected, the agent can book appointments directly into GoHighLevel, eliminating the need for third-party calendar tools.

Steps to Connect GoHighLevel

1. Navigate to Call Flows and locate the agent you want to integrate.
2. Ensure you are in the Advanced View.
3. Go to Integrations and click GoHighLevel.
4. Click the Connect button to authenticate and link your GoHighLevel account.
5. If you encounter issues, refer to the Troubleshooting section on the same page for guidance on common errors.
6. Once connected, go to Function Templates to view the prompt syntax required to activate the integration.
7. Copy the relevant syntax and paste it into the agent’s system prompt in the appropriate location.
8. Click Save to Draft or Publish to Production.

Key Notes

🎥 Watch the demo video below

27. Running Batch Calls

Overview

This guide explains how to set up and send a batch outbound calling campaign using an AI Voice Agent. Batch calls allow you to upload a list of contacts and have the agent call them automatically, with options to schedule campaigns and configure retry behaviour.

Steps to Run a Batch Call Campaign

1. Navigate to the Calls tab in the sidebar and open Batch Calls.
2. Select the agent you want to use for the campaign.
3. Confirm that the selected agent has a phone number assigned — this is required to send outbound calls.
4. Download the CSV template to see the required format for your contact list.
5. Prepare your CSV file with all contact numbers and any dynamic variables referenced in the agent’s prompt (e.g., first name, company).
6. Upload the CSV file.
7. Optionally, schedule the campaign to send at a specific date and time.
8. Configure Auto-Retry settings:
   • Set the number of retry attempts (e.g., up to 4 attempts)
   • Set the intervals between retries (e.g., retry after 3 hours, then 1 day, then 3 days)
9. Hover over the information icon (?) on each field for additional detail.
10. Launch the campaign.

Key Notes

🎥 Watch the demo video below

28. Changing Batch Call Speed

Overview

This guide explains how to configure the speed at which outbound batch calling campaigns are executed. Batch call speed is a workspace-level setting that applies to all agents within the workspace.

Steps to Change Batch Call Speed

1. Navigate to Settings in the sidebar.
2. Click Batch Call Settings.
3. Adjust the speed parameters — the calculated call rate at the bottom of the page will update in real time to reflect your changes.
4. Hover over the information icon (?) next to each field for a detailed explanation of how each parameter affects the call rate.
5. Click Save Settings.

Key Notes

🎥 Watch the demo video below

29. Managing Knowledge Bases

Overview

This guide explains how to create, manage, and attach knowledge bases to an AI Voice Agent. Knowledge bases provide the agent with reference information it can draw on during calls — such as business details, FAQs, pricing, and product information.

Steps to Create a Knowledge Base

Option 1 – From the Knowledge Base Tab

1. Open the sidebar and navigate to Knowledge Base.

2. Click Create New.

3. Choose the knowledge base type:

   Basic Knowledge Base

   • Enter the business name and key details
   • Expand the dropdown to add: opening hours, services offered, pricing, intake questions, FAQs, and other relevant information
   • Click Create when complete

   Advanced Knowledge Base

   • Upload a PDF, Word document, or text file, or paste raw text directly
   • Use this option for large or complex knowledge bases that exceed what the basic form supports
   • Click Create when complete

Option 2 – From Within an Agent

4. Navigate to Call Flows and locate the agent you want to edit.
5. Ensure you are in the Advanced View.
6. Go to the General tab and click Knowledge Base.
7. View currently attached knowledge bases, edit them, attach new ones, or detach existing ones.
8. Click Create New from this panel if you want to create a knowledge base directly from the agent.

Steps to Attach or Detach a Knowledge Base

1. Open the agent in Call Flows.
2. Ensure you are in the Advanced View.
3. Go to General → Knowledge Base.
4. Use the attach/detach controls to manage which knowledge bases are active for this agent.
5. Click Save to Draft or Publish to Production.

Key Notes

🎥 Watch the demo video below

30. Responsiveness & Interruption Sensitivity

Overview

This guide explains how to configure the speech detection settings for an AI Voice Agent, including how quickly the agent responds after the caller stops speaking, how sensitive it is to interruptions, and how it detects turn-taking in conversation.

Steps to Configure Speech Detection

1. Navigate to Call Flows and locate the agent you want to edit.
2. Ensure you are in the Advanced View.
3. Go to Voice and click Speech Detection.

Responsiveness

4. Adjust the Responsiveness slider — this controls how quickly the agent starts speaking after the caller stops talking. A higher value means the agent responds faster.

Interruption Sensitivity

5. Adjust the Interruption Sensitivity slider — this controls how many words the caller needs to say before the agent stops talking and yields the floor.
   • A higher setting means the agent will stop talking even for brief acknowledgements like “yes” or “okay”
   • A lower setting means the agent will continue speaking through short interjections and only yield when the caller says something more substantial — this reduces conversational overlap

Turn Detection Model

6. Select the Turn Detection Model:
   • STT-Based (Speech-to-Text) – uses speech-to-text processing to determine when the caller has finished speaking
   • AI Detection – uses AI to predict turn-taking

Preemptive Generation

7. Toggle Preemptive Generation on or off. When enabled, the agent begins generating its response before the caller has finished speaking. This can reduce latency but may result in incorrect responses if the caller’s full intent was not yet captured. It is generally recommended to leave this off.

8. Click Save to Draft or Publish to Production.

Key Notes

🎥 Watch the demo video below

31. Viewing Call History

Overview

This guide explains how to access and use the call history log in the portal. History provides a complete record of all calls made by your agents, including recordings, transcripts, post call analysis data, and call metadata.

Steps to View Call History

1. Open the sidebar and navigate to History.
2. The history log displays all calls across your workspace.
3. Click into any call record to view:
   • Recording – the audio recording of the call (if recording is enabled)
   • Transcript – the full text transcript of the conversation
   • Post Call Analysis – extracted data points configured in the Post Call Analysis settings
   • Post Call Task Statuses – the outcome status of any post-call actions (e.g., SMS sent, webhook fired)
   • Call Metadata – call ID, direction (inbound/outbound), source and destination numbers, Call Flow used, and call duration

Filtering the History Log

4. Use the filter options to narrow results by:
   • Date range
   • Specific agent
   • Functions called during the call
   • Call dispositions
   • Post Call Analysis data points (e.g., filter by a specific extracted value)
   • Call direction (inbound/outbound)
   • Batch campaign
   • Call status

Exporting History

5. Click Export to download the filtered history as a CSV file for use in spreadsheet tools or external reporting.

Key Notes

🎥 Watch the demo video below

32. Editing the Welcome Message

Overview

This guide explains how to configure the welcome message for an AI Voice Agent — the first thing the agent says (or does) when a call connects. Different message types are available for inbound and outbound calls.

Steps to Configure the Welcome Message

1. Navigate to Call Flows and locate the agent you want to edit.

2. Ensure you are in the Advanced View.

3. Go to the General tab and click Welcome Message.

4. Select whether you are configuring the Inbound or Outbound welcome message.

5. Set the delay — the number of seconds before the message is spoken after the call connects.

6. Toggle Interruptible on or off to control whether the caller can interrupt the welcome message before it finishes.

7. Select the message type:

   AI-Generated

   • Enter a prompt instruction (e.g., “Greet the caller and ask how you can help today”)
   • The agent will produce a slightly different variation of the greeting each time, based on the instruction

   Custom Message

   • Enter the exact wording you want the agent to say
   • The agent will deliver this message verbatim on every call
   • Dynamic variables can be included (e.g., “Hello {first_name}, thanks for calling”)

   Wait for Caller

   • The agent remains silent after the call connects and waits for the caller to speak first
   • Recommended for outbound calls only — on inbound calls, silence at the start of a call creates a poor caller experience
   • Suited to outbound scenarios where the recipient answers and says “Hello?” before the conversation begins

8. Click Save to Draft or Publish to Production.

Key Notes

🎥 Watch the demo video below

33. Integrating Google Calendar

Overview

This guide explains how to connect Google Calendar to an AI Voice Agent, allowing the agent to check availability and book appointments directly into a Google Calendar during a call.

Steps to Integrate Google Calendar

1. Navigate to Call Flows and locate the agent you want to integrate.
2. Ensure you are in the Advanced View.
3. Go to Integrations and click Google Calendar.
4. Click Connect with Google and follow the authentication prompts to link your Google account.
5. Once connected, refer to the Function Templates section below the connection panel — this contains the prompt syntax required to activate calendar functionality within the agent.
6. Review the syntax options to understand what each function does (e.g., check availability, create booking).
7. Copy the relevant syntax and paste it into the agent’s system prompt in the appropriate location.
8. Click Save to Draft or Publish to Production.

Key Notes

🎥 Watch the demo video below

34. Editing Voicemail Settings

Overview

This guide explains how to configure voicemail behaviour for an AI Voice Agent during outbound calls. When the agent encounters a voicemail, you can define whether it should hang up or leave a pre-configured message.

Steps to Configure Voicemail Settings

1. Navigate to Call Flows and locate the agent you want to edit.
2. Ensure you are in the Advanced View.
3. Go to Call Handling and click Voicemail Detection.
4. Hover over the information icon (?) for additional detail on how voicemail detection works.
5. Select the action to take when a voicemail is detected:
   • Hang Up – the agent ends the call immediately upon detecting voicemail
   • Leave Message – the agent records a message after the beep
6. If Leave Message is selected, enter the message text in the provided field. The agent will deliver this message verbatim.
7. Dynamic variables can be included in the message (e.g., “Hello {first_name}, I’m calling from…”) to personalise the voicemail using data from the contact list.
8. Click Save to Draft or Publish to Production.

Key Notes

🎥 Watch the demo video below

35. Editing Call Recording Settings

Overview

This guide explains how to configure call recording settings for an AI Voice Agent, including enabling or disabling recording and controlling whether recording links are publicly accessible.

Steps to Edit Call Recording Settings

1. Navigate to Call Flows and locate the agent you want to edit.
2. Ensure you are in the Advanced View.
3. Go to Security and click Recording.
4. Toggle Recording on or off to enable or disable call recording for this agent.
5. Toggle Public Recording Link on or off to control whether a publicly accessible URL is generated for each recording.
6. Click Save to Draft or Publish to Production.

Key Notes

Note: For more comprehensive security and compliance settings including HIPAA mode, data retention, and transcription redaction, refer to the Recording & Security Settings guide.

🎥 Watch the demo video below

KB Articles – AI Voice Portal (Batch 3)

36. Portal Overview

Overview

This guide provides a high-level walkthrough of the AI Voice Portal, covering the main sections and what each one is used for. Each feature area has a dedicated guide with more detailed instructions — this article is intended as an orientation for new users.

Dashboard

The Dashboard is the default landing page when you log in. It displays recently updated Call Flows and a summary of recent calls. All detailed work is done in the individual sections accessible via the sidebar.

Call Flows

Call Flows is where all agent configuration takes place. Open the sidebar and navigate to Call Flows to access your agents.

Within each Call Flow, configuration is organised across several tabs. Switch to the Advanced View to access the full range of settings.

General Tab

Voice Tab

Call Handling Tab

Data Tab

Security Tab

Security settings are particularly relevant for enterprise, medical, and legal use cases. Available settings include recording, compliance monitoring, HIPAA mode, transcription redaction, and data retention. See the dedicated Recording & Security Settings guide for full details.

Testing an Agent

To test an agent, click the Voice button to start a live voice session, or click Text to interact via text chat. Dynamic variables can be configured in the Variables panel before testing to simulate real call conditions.

A Logs panel is also available within the agent, showing all API call activity and executions for debugging purposes.

Sharing a Demo Link

To share an agent without needing a phone number, click Share Demo from the top of the agent editor. You can set an expiry date, per-user call limit, and total minute budget. Once created, the link can be shared with anyone to interact with the agent via a web browser. See the dedicated Sharing a Demo Link guide for full details.

Saving and Publishing

All changes can be saved in two ways:

If you make a mistake, use Prompt History within the Call Flow editor to review and revert to a previous version.

Knowledge Base

Navigate to Knowledge Base in the sidebar to create and manage knowledge bases independently of individual agents. There are two types:

• Basic – structured fields for business name, address, pricing, opening hours, FAQs, and intake questions
• Advanced – upload a PDF, Word document, or text file, or paste raw text directly

Once created, knowledge bases can be attached to agents from within the Call Flow editor. Advanced knowledge bases can be edited directly in the portal without re-uploading the source file. See the dedicated Managing Knowledge Bases guide for full details.

Phone Numbers

Navigate to Phone Numbers in the sidebar to purchase or bring your own numbers and assign them to agents. Select the country, search for an available number, claim it, and assign it to a specific Call Flow.

Calls

Navigate to Calls in the sidebar to manage outbound call campaigns.

See the dedicated Running Batch Calls guide for full details.

History

Navigate to History in the sidebar to view a complete log of all calls. Each record includes the recording, transcript, post call analysis data, and call metadata. Results can be filtered by agent, date range, disposition, call direction, batch, and post call analysis values. All data can be exported to CSV. See the dedicated Viewing Call History guide for full details.

Analytics

Navigate to Analytics in the sidebar to view call performance data across your workspace. Filter by agent or date range, view call volumes by time of day, manage dispositions, and export data to CSV. See the dedicated Viewing Analytics guide for full details.

🎥 Watch the demo video below

37. Providing Feedback on Test Calls

Overview

This guide explains how to provide feedback on test calls using the call ID and the shared feedback sheet. Structured feedback helps ensure that any issues with agent behaviour are addressed accurately and efficiently.

Steps to Provide Feedback

1. Navigate to History in the sidebar.
2. Locate the call you want to provide feedback on and click into it.
3. Find the Call ID at the top of the call record.
4. Click the Call ID — it will automatically copy to your clipboard.
5. Open the feedback sheet that has been shared with you.
6. Paste the Call ID into the relevant field.
7. Add your comments describing the issue.

How to Write Useful Feedback

Feedback is most actionable when it includes:

• What happened – describe the specific point in the call where the issue occurred
• What went wrong – explain what the agent said or did that was incorrect
• What should happen instead – provide the preferred response or behaviour

Feedback that only states something was wrong without providing context or the desired outcome is difficult to act on. Always include both the problem and the expected result.

Example of unhelpful feedback:

The agent asked the wrong questions.

Example of useful feedback:

At the point where the caller said they wanted to cancel, the agent asked for their account number before confirming whether cancellation was the right option. It should first confirm the intent and offer retention options before requesting account details.

Key Notes

🎥 Watch the demo video below

KB Articles – AI Voice Portal (Batch 4)

38. Integrating Cal.com

Overview

This guide explains how to integrate Cal.com with an AI Voice Agent to enable real-time calendar booking during calls. Cal.com connects with Outlook and Gmail calendars, allowing the agent to check availability and create bookings without any manual follow-up.

Before You Begin

You will need:

• A Cal.com account with at least one event configured
• A Cal.com API key
• The Event Type ID for the event you want to use

Part 1 – Set Up Cal.com

Create an Event

1. Log into your Cal.com account.
2. Create a new event and give it a name (e.g., Consultation, Site Visit).
3. Add a description if required and set the event duration.
4. Click Continue.
5. Configure additional settings as needed:
   • Location – in-person, Microsoft Teams, or other options
   • Availability – set the windows during which bookings can be made
   • Limits – cap the number of bookings per day or week if required

Get the Event Type ID

6. Once the event is saved, locate the Event Type ID — this is the numeric code shown in the URL or event settings.
7. Copy this ID — you will need it in the next step.

Generate an API Key

8. In Cal.com, navigate to Settings → Developer → API Keys.
9. Click Add New API Key and give it a name.
10. Set an expiration date or select Never Expire.
11. Click Save — your API key will be generated and displayed once.
12. Copy the API key immediately and store it securely.

Part 2 – Connect Cal.com to the Agent

1. Navigate to Call Flows and locate the agent you want to integrate.
2. Ensure you are in the Advanced View.
3. Go to Integrations and click Cal.com.
4. Enter the following details:
   • API Key – paste the API key generated in Cal.com
   • Event Type ID – paste the numeric ID from the Cal.com event
5. Select the Time Zone relevant to the bookings.
6. Optionally, enter an Availability ID if you have configured specific availability windows in Cal.com.
7. Enter a Text to Speak message — this is what the agent will say while the booking function is being executed (e.g., “Just give me a moment while I check availability for you”). This prevents silence during the lookup.
8. If you want additional information captured during the call to be included in the calendar booking (e.g., a caller’s address for a site visit), map the relevant dynamic variables to the corresponding Cal.com booking fields.

Part 3 – Update the System Prompt

9. Go to Function Templates within the Cal.com integration panel — this contains the prompt syntax required to activate the Cal.com functions.
10. Copy the relevant syntax and paste it into the agent’s system prompt at the appropriate point in the call flow.
11. The prompt should define when the agent checks availability and when it creates a booking.
12. Click Save, then click Save to Draft or Publish to Production.

Key Notes

🎥 Watch the demo video below

39. Connecting Your Business Number (Call Forwarding)

Overview

This guide explains how to configure call forwarding from your existing business mobile to your AI Voice Agent. Once set up, calls to your number can be automatically handled by the agent based on the conditions you define — such as when you are busy, unavailable, or for all calls.

Before You Begin

• The agent must have a phone number assigned to it before connecting your business number.
• You will need access to your mobile phone to dial the forwarding codes or scan the QR code provided.

Steps to Set Up Call Forwarding

1. Navigate to Call Flows and locate the agent you want to connect.
2. Ensure you are in the Advanced View.
3. Click Connect to Business.
4. Select your country (e.g., Australia).
5. Select your carrier from the list — the three major carriers are shown by default, with additional carriers available by expanding the list.
6. Click Continue.
7. Select your device type and click Continue.
8. Choose the forwarding condition — select the scenario in which calls should be forwarded to the AI agent:

9. Click Continue.
10. Optionally, toggle Disable Voicemail / Message Bank if you want to turn off your carrier voicemail — recommended to avoid calls going to voicemail before reaching the agent.
11. Follow the on-screen instructions — either scan the QR code or dial the provided codes on your mobile to activate forwarding.
12. Click Done.
13. Click Save to Draft or Publish to Production.

Key Notes

🎥 Watch the demo video below

40. Managing Phone Numbers

Overview

This guide explains how to purchase a phone number and assign it to an AI Voice Agent. Phone numbers are required for agents to make or receive calls.

Steps to Purchase and Assign a Phone Number

Purchase a Number

1. Open the sidebar and navigate to Phone Numbers.
2. Click Buy a Number.
3. Select the country you want the number to be based in.
4. Click Search — available numbers will be displayed.
5. Select a number and click Claim to purchase it. The number will be allocated to your account.

Configure the Number

6. Click into the newly purchased number to open its settings.
7. Enter a nickname to help identify the number within the portal.
8. Select the agent you want to assign this number to.
9. Configure any additional settings as required.
10. Click Save Configuration.

Key Notes

🎥 Watch the demo video below

41. Integrating Slack

Overview

This guide explains how to connect Slack to the portal and enable Slack notifications for an AI Voice Agent. Once configured, the agent will send a notification to your Slack workspace at the end of each call.

Steps to Integrate Slack

Step 1 – Connect Slack at the Workspace Level

1. Navigate to Settings in the sidebar.
2. Click Slack and follow the prompts to connect your Slack workspace.

Step 2 – Enable Slack Notifications on an Agent

3. Navigate to Call Flows and locate the agent you want to configure.
4. Ensure you are in the Advanced View.
5. Go to Data and click Slack Integration.
6. Toggle Slack Notifications on to enable notifications for this agent, or off to disable them.
7. Click Save to Draft or Publish to Production.

Key Notes

🎥 Watch the demo video below