Call Control

Speaker Diarization

Speaker Diarization allows the system to identify and separate different speakers during a call. When enabled, the agent can detect when multiple people are speaking and label them individually in the call logs.

How It Works

• When Speaker Diarization is turned on, the system analyzes the audio and assigns labels such as User 1, User 2, etc.

• In the Call Logs, you will clearly see which speaker said what, making multi-speaker conversations easier to review and understand.

Ignore Secondary Speakers

The Ignore Secondary Speakers option works together with Speaker Diarization.

When enabled:

• The system will still display the text from additional speakers (e.g., User 2, User 3) in the logs.

• However, the agent will only respond to the primary speaker and will ignore all secondary speakers.

• This prevents the agent from getting confused or responding to background conversations.

Greeting Message Configuration

The Greeting Message Configuration section in Call Control allows you to manage how the agent begins a voice interaction. It includes options for controlling when the agent starts speaking and whether the user can interrupt the greeting.

1. Wait for Greeting

When Wait for greeting is enabled:

• The agent will wait for the user to start speaking before responding.

• You can specify a wait duration (e.g., 5 seconds).

• If the user does not speak within the configured time, the agent will automatically start its greeting message.

This setting is helpful when you want the user to initiate the conversation or when you want to avoid the agent speaking too early.

2. Block Interruptions

The Block Interruptions option (Only applies in Custom Mode) works together with the Custom Greeting Message.

When Block Interruptions is enabled:

• The agent will always begin the call with the custom greeting message you wrote, regardless of the selected language.

• During this greeting, the user cannot interrupt the agent—speech input will be ignored until the greeting is finished.

• This ensures the greeting is delivered fully and clearly every time.

This option is especially useful when the greeting contains important information or instructions that the user must hear before interacting with the agent.

Blocking Interruptions During the Call

In addition to blocking interruptions during the greeting message, you can also prevent users from interrupting the agent at any time during the conversation.

To do this, enable Block Interruptions in the Interruptions section.

How It Works

• When Block Interruptions is enabled in the Interruptions section:

• The agent will continue speaking without being cut off by the user.

• Any user speech that happens while the agent is talking will be ignored until the agent finishes its response.

• This setting ensures the agent can deliver full messages or instructions without disruption.

This feature is useful for scenarios where the agent must provide complete information before taking user input, such as compliance statements, instructions, or important announcements.

Timeouts

The Timeouts settings allow you to control how long the agent waits for user responses and how long a call or text conversation should last. These settings help ensure smooth interactions and prevent calls from staying open unnecessarily.

1. Node Repeat Timeout

• This defines how many seconds the agent will wait for a user response before repeating its last message.

• If the user does not respond within the specified time, the agent will repeat the prompt.

• The minimum allowed value is 30 seconds.

Example: If the timeout is set to 30 seconds, the agent will repeat its question every 30 seconds until the user responds.

2. End Call Timeout

• This is the amount of idle time (no user speech) the agent will wait before ending a voice call entirely.

• If the user does not speak within the specified time, the call automatically ends.

Example: If set to 60 seconds, the agent will end the call if the user stays silent for 60 seconds.

3. End Text Call Timeout

• Works the same as End Call Timeout, but applies to text-based calls instead of voice calls.

• If the user does not send any message within the configured time, the text session ends.

4. Max Voice Call Duration

• This setting defines the maximum total length of the entire voice call, in seconds.

• When the limit is reached, the system will end the call automatically.

Example: If set to 600 seconds, the call will end after 10 minutes, regardless of activity.

VoIP Call Back Settings

The VoIP Call Back Settings allow you to automate situations where the agent should call the user back after a call fails or ends under specific conditions. This feature ensures that important calls are not missed and the user receives timely follow-ups.

1. Callback Trigger Conditions

You can configure the agent to call the user back depending on the call outcome. The available conditions include:

• Busy – The user’s line was busy.

• Unanswered – The user did not pick up the call.

• Dropped – The call disconnected unexpectedly.

• Voicemail – The call went to voicemail.

When any of the selected conditions occur, the callback workflow will start automatically.

2. Callback Delay

You can choose how long the agent should wait before calling back. The options include:

• Minutes

• Hours

• Days

This allows you to set an appropriate follow-up delay, such as calling back after 5 minutes, 1 hour, or 1 day, depending on your use case.

3. Number of Callback Attempts

You can also specify how many times the agent should attempt to call the user back.

• For example, setting this to 3 means the agent will try up to three callbacks based on your configured conditions and delays.

Limiting the number of attempts prevents excessive callbacks and ensures a controlled, predictable user experience.