​ To learn more, see the sources I considered: [[1] MCP's Speak During Execution is not working.](https://discord.com/channels/1186424376300679209/1411987278842298409) [[2] Prompt Engineering Guide](https://docs.retellai.com/build/prompt-engineering-guide) [[3] Multiple tool calls invoked while an agent is still reading the response](https://discord.com/channels/1186424376300679209/1429919905179893974) ​

### Answer The "Speak During Execution" feature triggers a single statement at the start of tool execution—it doesn't continuously fill silence while tools are running. [(1) ]() For your scenario with multiple sequential tool calls, there isn't a built-in mechanism to speak only once across all tool calls without configuring each tool individually. If you enable "Speak During Execution" on each tool, it would indeed speak before each one. Workarounds to consider: 1. Use prompt engineering to add a brief "working…" or "give me a moment" message in your pre-execution prompt before triggering the tools. [(2) ]() 2. Use "Speak After Execution" to immediately confirm results once all tools finish. [(1) ]() 3. Consolidate tool calls where possible—if multiple tools are being triggered together, consider whether they can be combined into a single custom function that handles all the logic server-side, reducing the number of individual tool invocations. 4. Adjust your prompting to explicitly instruct the agent to say something first, then execute tools. For example: "call this function ONE time then say '…'" [(3) ]() The documentation doesn't describe a specific setting for handling cumulative silence across multiple parallel tool calls. You may want to reach out for additional guidance: - Visit [Contact Sales]() - Check [) Would you like a general answer based on my knowledge instead?

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