### Answer --- Transitions in Conversation Flow happen when the user is done speaking or when Skip Response is enabled and the agent finishes speaking. [(1) ]() For your scenario of "press 1 or hold while we look for a human": Option 1: Use a Conversation Node with Skip Response If you want the agent to say something like "Press 1 or hold while we transfer you" and then automatically transition without waiting for user speech, you can enable Skip Response on the node. When enabled, the agent will transition to the next node via a single edge after it finishes speaking. Option 2: Use Prompt-based Transition Conditions You can define transition conditions using prompts such as: -

User said they want to hold

-

User said they want to speak to a human

-

User pressed 1

[(2) ]() Important limitation with Press Digit nodes: Be aware that Press Digit nodes only trigger transitions **when the user is done speaking**—the digit press alone does not create a dialogue turn that advances the flow. [(3) ]() This means if you're relying solely on DTMF input without any speech, the flow may stall. Workaround: Combine the hold/transfer scenario with a Conversation Node that prompts the user to speak or press a digit, then use prompt-based conditions to detect either action. [(2) ]()

​ To learn more, see the sources I considered: [[1] Conversation Node](https://docs.retellai.com/build/conversation-flow/conversation-node) [[2] Step 3: Add transition conditions](https://docs.retellai.com/build/conversation-flow/transition-condition) [[3] 🚨 press_digit does not trigger node transitions — flow stalls unless user speaks](https://community.retellai.com/t/press-digit-does-not-trigger-node-transitions-flow-stalls-unless-user-speaks/627) ​

Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.

@AI Evy can you help me?

### Answer --- You're correct—if you use Skip Response with a single edge, the flow will transition directly to that destination node after the agent finishes speaking, and there's no opportunity for the user to interrupt and go somewhere else based on pressing a digit. [(1) ]() The core challenge: Transition conditions in Conversation Flow are evaluated when the user is done speaking. Press Digit nodes specifically do not create a dialogue turn on their own—the digit press alone does not advance the state machine unless the user also speaks. [(2) ]() Possible approach: Instead of using Skip Response, keep the node as a standard Conversation Node with **multiple transition conditions**: - One prompt-based condition like:

User pressed 1

or

User wants to speak to a human

- Another prompt-based condition like:

User said they want to hold

or

User stayed silent

This way, if the user interrupts and says something (or presses a digit and the IVR/system acknowledges it audibly) , the transition conditions can evaluate and route accordingly. [(1) ]() Limitation to be aware of: Based on community reports, DTMF-only scenarios (where the user presses a digit but never speaks) may not trigger transitions reliably, as the platform evaluates transitions on speech input. [(2) ]() If your use case requires detecting a digit press mid-locution without any speech, this may not be fully supported at this time. You may want to reach out to Retell support directly for guidance on this specif

ic scenario or to confirm if there have been any recent changes to this behavior.

​ To learn more, see the sources I considered: [[1] Step 3: Add transition conditions](https://docs.retellai.com/build/conversation-flow/transition-condition) [[2] 🚨 press_digit does not trigger node transitions — flow stalls unless user speaks](https://community.retellai.com/t/press-digit-does-not-trigger-node-transitions-flow-stalls-unless-user-speaks/627) ​

Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.

### Answer --- This is a challenging scenario because of how transitions work in Conversation Flow. Transitions typically happen when the user is done speaking. [(1) ]() The core limitation: For a "hold and wait silently" scenario where the user says nothing, the platform evaluates transitions based on speech input. A digit press alone does not create a dialogue turn that advances the flow. [(2) ]() Potential workaround using `begin_after_user_silence_ms`: You could potentially use the

begin_after_user_silence_ms

setting. If the agent is configured to wait for the user to speak first, this setting specifies how long (in milliseconds) the AI will wait before beginning the conversation. [(3) ]() However, this applies at the conversation flow level for when the agent should start speaking, not for mid-conversation silence-based transitions. What you might try: 1. Use a timeout-based approach: After your initial prompt ("press 1 or hold") , if the user remains silent for a period, some implementations handle this by having the IVR/system eventually produce audio (like hold music or a "please continue to hold" message) , which would then trigger the transition evaluation. [(4) ]() 2. Restructure the flow: Instead of relying on silence detection, you could prompt the user to say "hold" or "wait" as an alternative to pressing 1, since prompt-based conditions like

User said they want to hold

can be evaluated. [(1) ]() Unfortunately, based on the available

documentation, there isn't a native "transition on user silence" condition for mid-conversation scenarios. This appears to be a platform limitation for DTMF-only or silence-based flows. [(2) ]() I'd recommend reaching out to Retell support directly to confirm if there's a workaround or upcoming feature for this use case.

​ To learn more, see the sources I considered: [[1] Step 3: Add transition conditions](https://docs.retellai.com/build/conversation-flow/transition-condition) [[2] 🚨 press_digit does not trigger node transitions — flow stalls unless user speaks](https://community.retellai.com/t/press-digit-does-not-trigger-node-transitions-flow-stalls-unless-user-speaks/627) [[3] Create Conversation Flow](https://docs.retellai.com/api-references/create-conversation-flow) [[4] Press Digit Node](https://docs.retellai.com/build/conversation-flow/press-digit-node) ​

Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.

### Answer You can contact Retell AI support through these channels: - **Email**: support@retellai.com [(1) ]() - **Contact Sales**: [Contact Sales]() [(2) ]() - **GitHub**: [) When reaching out, it's helpful to include your Workspace ID (found in account settings at the bottom left of the dashboard) so the team can investigate your specific account.

​ To learn more, see the sources I considered: [[1] Ned help from Support](https://community.retellai.com/t/ned-help-from-support/429) [[2] Verified Phone Number Support](https://community.retellai.com/t/verified-phone-number-support/432) ​

Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.

### Answer No, based on the available documentation, there is no configurable delay or wait time before a Logic Split Node evaluates its conditions. [(1) ]() When entering a Logic Split Node, the agent immediately evaluates the conditions and branches out to the corresponding destination nodes. The agent does not speak in this node, and the time spent in this node is minimal. The Logic Split Node is designed to be a split point that evaluates conditions instantly—it's not intended to introduce any waiting period. For your use case (wanting to wait for user input like pressing 1 or holding before branching) , a Logic Split Node isn't the right tool. Logic Split Nodes are meant for branching based on dynamic variables or conditions that are already known, not for waiting on user actions. You may want to reach out to Retell support at support@retellai.com to discuss whether there's a different approach or upcoming feature that could support a timeout-based transition for your "press 1 or hold" scenario.

​ To learn more, see the sources I considered: [[1] Logic Split Node](https://docs.retellai.com/build/conversation-flow/logic-split-node) ​

Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.

### Answer Based on the available documentation, there is no built-in "wait time" or delay setting that you can configure before transitioning to another node in Conversation Flow. [(1) ]() , [(2) ]() The Logic Split Node evaluates conditions immediately when the agent enters it—the agent does not speak, and time spent in this node is minimal. [(1) ]() The only time-related setting available is

begin_after_user_silence_ms

, which specifies how long the AI will wait for the user before beginning the conversation. However, this only applies if the agent is configured to wait for the user to speak first at the start of the conversation—not for mid-conversation transitions. [(3) ]() For your "press 1 or hold" scenario, there isn't a native way to add a wait/timeout before transitioning. This appears to be a limitation of the current platform. Check out these potentially useful sources: - [Logic Split Node]() - [Create Conversation Flow]() - [Node Overview]() I'd recommend reaching out to Retell support at support@retellai.com to discuss whether a timeout-based transition feature is available or planned.

​ To learn more, see the sources I considered: [[1] Logic Split Node](https://docs.retellai.com/build/conversation-flow/logic-split-node) [[2] Node Overview](https://docs.retellai.com/build/conversation-flow/node) [[3] Create Conversation Flow](https://docs.retellai.com/api-references/create-conversation-flow) ​

Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.

### Answer Based on the available documentation, there isn't a native way to have a Conversation Node "wait" silently without the agent speaking. What the documentation shows: - Call Transfer Node with warm transfer has an on-hold music option—audio played to the caller while they are on hold. [(1) ]() - This is the only documented mechanism for having the caller wait without the agent speaking, but it's specifically tied to the transfer process. For your "press 1 or hold" scenario: One potential approach would be to use a Call Transfer Node with warm transfer configured with on-hold music. The caller would hear the hold music while waiting for the transfer destination to connect. However, this doesn't solve the mid-conversation "wait silently then branch" scenario you're describing—it's designed for actual call transfers. The available sources don't document a way to configure a Conversation Node to remain silent for a set duration while waiting for user input (like a digit press or continued silence) . Check out these potentially useful sources: - [Call Transfer Node]() - [Node Overview]() - [Create Conversation Flow]() I'd recommend reaching out to Retell support at support@retellai.com to confirm if there's a workaround or feature request process for this use case.

​ To learn more, see the sources I considered: [[1] Call Transfer Node](https://docs.retellai.com/build/conversation-flow/call-transfer-node) ​

Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.