Here’s an overview of our VoiceAgent capabilities and how they can be configured to enhance and automate your communication workflows.
User Level:
Admin
Introducing VoiceAgents at CloudTalk
CloudTalk VoiceAgent is an AI-powered virtual agent designed to handle voice interactions autonomously, helping businesses scale and streamline their call workflows. Whether operating as the first point of contact or running fully automated conversations, VoiceAgents enable teams to offload repetitive tasks and ensure consistent, high-quality caller experiences.
With VoiceAgents, you can automate a wide range of tasks, such as:
Reaching out to leads with personalized outbound calls
Handling onboarding calls or surveys without tying up human agents
Collecting structured data during calls to feed into CRMs or internal systems
Answering FAQs or intelligently routing callers to the right department or person
VoiceAgents are highly customizable, integrating with your existing workflows, tools, and data sources to deliver efficient, context-aware conversations - at scale.
Note: VoiceAgent functionality is currently not available for all accounts.
If you do not see the VoiceAgents tab in your CloudTalk Dashboard and would like to explore this feature, please contact our Support team or your Customer Success Manager for assistance.
Billing Information
VoiceAgent calls are billed on a per-minute basis. Depending on your billing configuration, charges are either deducted directly from your CloudTalk balance or reflected in your monthly invoice.
💡To manage usage effectively, you can define a maximum call duration per agent. Once this limit is reached, the call will automatically terminate to prevent additional charges.
Setting Up and Configuring a VoiceAgent
To manage your VoiceAgents, go to Dashboard > VoiceAgents > Agents. From this section, you can view all existing agents or initiate the creation of a new one.
2. Click the ➕ (Add) icon to begin creating a new VoiceAgent
3. Enter the general information for your VoiceAgent to define its identity and purpose.
Setting | Description |
Agent Name | The internal name of the VoiceAgent. This name will appear in Call Logs, Analytics, and any Integrations (e.g., in HubSpot as the Agent Name). Choose a name that clearly identifies the agent’s purpose or target audience (e.g., "Demo Booker", "Onboarding Agent"). |
Language | Currently, only English is supported. |
Call Direction | You can configure your VoiceAgent to handle either inbound or outbound calls, depending on your use case. Inbound VoiceAgents are ideal for answering incoming calls, providing information, or qualifying callers before transferring them to a human agent. Outbound VoiceAgents are great for reaching out to leads, following up on tasks, or booking meetings. When setting up an inbound VoiceAgent, you also have the option to define a maximum call duration. This lets you control how long the AI stays on the call. If the limit is reached, the call will end automatically, even if the conversation is still ongoing. |
Outbound Number | Choose a specific number or set to Automatic. If set to Automatic, the system will use a number that perfectly matches the destination country. For example, if calling a German number and a German number exists in your account, it will be used. However, the system does not pick the "closest" alternative - only exact country matches are supported. If no match is found, the Failover Number will be used. |
Failover Number | Backup number used if no local number is available. |
Maximum Call Duration | Add a hard limit (in minutes) to control costs. When the limit is reached, the call is immediately terminated - even mid-sentence. The VoiceAgent will not attempt to wrap up or finish its task. We recommend setting a buffer above the expected conversation length to avoid abrupt call endings. |
4. Set up the Conversation Behavior parameters for your VoiceAgent. This includes defining the VoiceAgent Prompt, where you establish the agent’s persona and the conversational context it will follow during calls. These settings help shape how the agent interacts with callers, ensuring alignment with your desired tone, style, and communication goals.
Want help writing strong prompts? See our Prompt Writing Guide.
5. If required, activate the VoiceAgents Call Transfers setting to allow your VoiceAgent to redirect calls to live agents. Once enabled, you can define the specific conditions that trigger a transfer.
For optimal performance of the call transfer function, we recommend avoiding the use of the GPT-4o-mini model, as it has shown inconsistent behavior. Instead, we suggest using either the GPT-4.1-mini model, which offers similar latency with improved reliability, or the full GPT-4o / GPT-4.1 models for more advanced capabilities and stable transfer handling.
6. Configure Data Capture & Output settings. Set up Call Analysis Prompts to define what data to extract from call transcripts, and specify a Call Results Endpoint to determine where that data should be sent. This enables automated updates to CRMs, workflows, and analytics tools. Learn more about configuring Data Capture & Output configuration in dedicated article.
7. Configure Voice & Model Settings. Choose your AI provider, model, and voice settings:
Setting | Description |
Provider | Deepgram or Elevenlabs |
Language Model | OpenAI or Anthropic. Choose based on use cases: Fast/simple replies → GPT-4o Mini; Deep, conversational → GPT-4o / Claude Sonet; Structured & concise → Claude Haiku. |
Voice | Pick your VoiceAgent’s voice. You can preview options directly in the dashboard. Some voices include gender, accent, and tone labels (e.g., Jessica - young, American, female) |
Temperature (Elevenlabs only) | Controls response style. Low (0-0.3) = more factual and robotic, High (0.6 - 1) = more expressive and varied. Useful for adjusting the "personality" of the voice. |
Streaming Latency (Elevenlabs only) | Controls response speed. Lower latency = faster response time, but may reduce quality. Higher = smoother output, but with more delay. Low latency may reduce naturalness slightly. |
Stability (Elevenlabs only) | Higher = more consistent tone; good for professional agents. Lower = more dynamic/emotional; great for sales or casual calls. |
Similarity (Elevenlabs only) | Controls how closely the voice mimics its reference; higher = clearer, more exact match, but risks sounding synthetic. Lower = warmer and more fluid, but slightly less consistent. |
8. Select and assign a Knowledge Base to your VoiceAgent to extend the information it can reference during calls. This enables the agent to deliver more accurate and context-aware responses based on your curated content.
9. Configure Tools to allow your VoiceAgent to interact with third-party applications in real time. This enables dynamic actions during calls—such as retrieving data, updating records, or triggering workflows—based on the conversation context.
10. Save VoiceAgent configuration.
Triggering VoiceAgent Calls
CloudTalk VoiceAgents can be used in two different ways, depending on when you want the AI to interact with your customers:
Inbound Calls
VoiceAgent can handle incoming calls as part of your call flow. When a customer calls your number, the VoiceAgent can take over the conversation at a specific point in the flow. It can help with answering questions, checking order or ticket status, booking appointments, and more.
To learn how to set this up, see the VoiceAgent for Inbound Calls guide.
Outbound Calls
You can also use VoiceAgent to make outgoing calls. These calls are triggered automatically using the CloudTalk API, often through tools like HubSpot, Zapier, Make, or your internal systems. This is useful for tasks like follow-ups, reminders, surveys, or proactive outreach without needing a human agent to start the call.
To learn more, see the VoiceAgent for Outbound Calls guide.
If you need further assistance or have any questions, you can contact our Support team. We are always here to help you!