Voice Steps

What is a Voice Step?

A Voice Step represents a conversational turn:

Assistant speaks — The configured script is spoken via text-to-speech
User responds — The system listens and transcribes the response
AI evaluates — The LLM analyzes the response against defined conditions
Navigation — The conversation moves to the appropriate next step

Voice Step Structure

Property	Description
Script	What the assistant says
Conditions	Rules for navigation based on the response
Prompt Mode	Verbatim (word for word) or Instruct (an instruction the AI follows). See Writing Scripts.
Response Eagerness	Response timing: Keen, Normal or Patient
Thinking Effort	Reasoning depth: Fast, Normal or Deep
Silence Tolerance	How to handle caller silence
Interruptions	Whether the caller can talk over the line, and the Verbatim threshold. See Interruptions.
Fast Response	Whether the step waits for a reply. See Fast Response.
Models	Which AI model writes the reply, in Instruct mode. See Models.
Key Terms	Words the recogniser should listen for while this step runs, and the Overfit option. See Key Terms.
Classifiers	Background extractions that run alongside the step, defined for this step or shared across the script or flow. See Classifiers.

All of these are set from the step’s menu. See Step-Specific Settings for the options and defaults.

Writing Scripts

The script defines what your assistant communicates. The step’s Prompt Mode (on the step menu) decides how it is used:

Prompt Mode	What happens
Instruct (default)	The script is an instruction. The assistant works out what to say, in its own words, from the script and the conversation so far.
Verbatim	The script is spoken exactly as written, with no changes by the AI.

New steps start on Instruct. We recommend Verbatim for any line you can write in advance — greetings, disclosures, read-backs and closing lines — and Instruct for turns that have to respond to what the caller just said. Older steps can show a Say badge from an earlier prompt mode. They run as Instruct steps told to say the text, so the wording can vary slightly. Switch them to Verbatim for exact wording or Instruct for guided wording.

Verbatim Scripts

Choose Verbatim when the wording must be exact: a disclosure, a legal statement, a greeting you’ve tuned carefully. Write the line exactly as it should be heard:

Hi, this is Sarah from Voxworks. Is now a good time to chat?

If the caller cuts in before a verbatim line has reached the step’s Verbatim threshold (80% unless you change it), the line is said again word for word when the step runs again; past the threshold it counts as said and the assistant carries on in its own words. See Interruptions.

Instructional Scripts

With Instruct, use describing words like “consider”, “ask about”, or “explain” for flexible delivery:

Explain the benefits of our platform and gauge their interest.
Ask about their availability for a follow-up call.

The assistant will interpret instructional scripts and generate natural responses that convey the intended meaning.

Tips for Scripts

Keep sentences short and clear
Ask one question at a time
Include natural transitions

Conditions

Conditions determine where the conversation goes based on the user’s response:

Step: "Would you like to schedule a call this week?"
   Condition: "User agrees or shows interest"
   → Next: booking_step
   → Script: "Perfect! Let me check what times are available."

   Condition: "User wants more information first"
   → Next: info_step
   → Script: "Of course! What would you like to know?"

   Condition: "Otherwise"
   → Next: objection_step
   → Script: "I understand. Can I ask what's holding you back?"

How Conditions Are Evaluated

The AI doesn’t do simple keyword matching. Instead, it:

Understands context — Reads the caller’s reply alongside what the assistant said on this step
Interprets intent — Determines what the user actually means
Matches conditions — Selects the most appropriate condition
Generates response — Creates a natural response incorporating the next script

This means users can express the same intent in many ways:

User Says	Matched Condition
”Yes, let’s do it”	User agrees
”Sure, sounds good”	User agrees
”I’m interested”	User agrees
”Tell me more first”	User wants more info
”What does it cost?”	User wants more info
”Not right now”	Otherwise

Special Conditions

Otherwise

The “otherwise” condition is a fallback when no specific condition matches:

Always include an “otherwise” condition
Use it to handle unexpected responses gracefully
On a question, point it at a step that asks again in different words; on a statement that should move straight on, point it at the next step

Ending the call from a condition

As well as a next step, a condition can end the flow or end the call. End Call lets the assistant finish speaking and still leaves the caller a short window to keep the call going; End Call (Strict) finishes the sentence and hangs up. See Where a condition can send the call.

Question Answering

If the user asks a question that can be answered from the Knowledge Base or context, the assistant can respond without leaving the current step. This is handled automatically — the assistant answers the question and then continues with the current step’s flow.

Interrupts

By default (Normal in the step’s Interruptions menu), the first few seconds of each line are protected, and after that the caller can speak over the assistant, causing it to stop and process the interruption. You can change this per step:

Uninterruptible: the line always plays in full. Anything said over it is answered afterwards.
Uninterruptible turn: nothing the caller says during the step’s turn is acted on, though it is still recorded.
Allow skip: if the caller speaks before the line starts, the line is dropped.

See Interruptions for when to use each. When an interrupt occurs on a Normal step, if the assistant believes it hasn’t fully conveyed its message for the step, it will attempt to repeat the key information before continuing. This ensures important details aren’t lost due to interruptions. See Assistant Behaviour for more on how the assistant handles interruptions.

Best Practices

One question per step — Don’t overload the user
Anticipate responses — Think about all ways users might reply
Include fallbacks — Every voice step ends with an “Otherwise” (the editor won’t save one that doesn’t). On a question, point it at a step that asks again in different words; on a statement that should move straight on, point it at the next step.
Keep transitions natural — Next steps should flow from the response
Test with real language — Users don’t speak in keywords

Next Steps

Steps Overview — Learn about all step types
Conversation Dynamics — Fine-tune timing and response behavior
Variables — Use data in your scripts

Get Started & Build

Calls, Contacts & Batches

Numbers & Routing

Tools, Automations & Integrations

Account, Billing & Security

What is a Voice Step?

Voice Step Structure

Writing Scripts

Verbatim Scripts

Instructional Scripts

Tips for Scripts

Conditions

How Conditions Are Evaluated

Special Conditions

Otherwise

Ending the call from a condition

Question Answering

Interrupts

Best Practices

Next Steps

​What is a Voice Step?

​Voice Step Structure

​Writing Scripts

​Verbatim Scripts

​Instructional Scripts

​Tips for Scripts

​Conditions

​How Conditions Are Evaluated

​Special Conditions

​Otherwise

​Ending the call from a condition

​Question Answering

​Interrupts

​Best Practices

​Next Steps

What is a Voice Step?

Voice Step Structure

Writing Scripts

Verbatim Scripts

Instructional Scripts

Tips for Scripts

Conditions

How Conditions Are Evaluated

Special Conditions

Otherwise

Ending the call from a condition

Question Answering

Interrupts

Best Practices

Next Steps