HomeCasa Prompts

IVR phone menu voice prompts ready for ElevenLabs

Design your phone menu tree and get every prompt (welcome, options, after-hours, hold, invalid input) as a separate audio file, with numbers normalized for AI text to speech. Designed for ChatGPT, Claude, Gemini and ElevenLabs. It includes 6 customizable variables and you can copy it for free in one click.

Intermediate · Audio · Pending editorial review

Designed for:

Prompt

Act as a voice experience designer for customer phone lines. Design my IVR phone menu and write every voice prompt ready to generate with ElevenLabs.

Details:
- Company: {{company}}
- Industry: {{industry}}
- Most common call reasons, most to least frequent: {{reasons}}
- Business hours: {{hours}}
- Menu languages: {{languages}}
- Brand tone: {{tone}}

Deliverables:
1. Menu tree: at most 4 options per level and 2 levels deep, most frequent first, with a way to reach a person always available.
2. A list of prompts, each as its own file with a suggested name (e.g. 01_welcome.mp3): welcome, main menu, submenus, after-hours, holidays, hold message, invalid input, no input and transfer.
3. Text of each prompt prepared for AI voice: numbers, times and emails written the way they should be spoken; sentences under 20 words; action before the key (“To check an order, press 1”).
4. Pauses: insert <break time="1.0s" /> between options (works with ElevenLabs Multilingual v2 and Flash models; with Eleven v3, use ellipses instead).
5. Estimated length of each prompt and a voice recommendation (register, speed, style).

Constraints:
- Do not invent phone numbers, web addresses, hours or timelines: use [FILL IN].
- If calls are recorded, include a recording and privacy notice marked [REVIEW WITH LEGAL COUNSEL].
- If information is missing to design the tree, ask me first.

Format: the tree as a nested list and a prompt table (file | text | length).

How to use it

  1. Rank call reasons by real frequency (check your call logs if you have them).
  2. Generate each prompt separately in Text to Speech with the same voice and settings, and download them with the suggested names.
  3. Upload the files to your phone system and check which audio format it accepts.
  4. Make test calls and shorten any message that drags.

Usage example

Company: Bright Smile Dental · Industry: healthcare · Reasons: book, reschedule, billing, emergencies · Hours: Mon-Fri 8 am-6 pm · Languages: English and Spanish · Tone: warm and calm.

Sample output

Description of the generated audio (what you would hear), not a recording.

Prompt structure

FileApprox. lengthWhat you hear
01_welcome.mp35 swarm, calm female voice, unhurried
02_language.mp35 s"For English, press 1. Para español, oprima 2."
03_recording_notice.mp36 scall recording and privacy notice [REVIEW WITH LEGAL COUNSEL]
04_main_menu.mp318 sfour options separated by 1 s pauses, action before key
05_after_hours.mp312 shours spoken naturally plus the emergency option
06_hold.mp38 sfriendly hold message, looped over soft music

Text sample

"Thank you for calling Bright Smile Dental." "To book an appointment, press one. <break time="1.0s" /> To reschedule or cancel, press two. <break time="1.0s" /> For billing questions, press three. <break time="1.0s" /> For a dental emergency, press four. <break time="1.0s" /> To speak with our team, press zero." "Our office is open Monday through Friday, from eight a.m. to six p.m."

Notes

For a serious medical emergency, the emergency prompt tells callers to hang up and call [FILL IN: emergency number]. Break tags work with Multilingual v2 and Flash; with v3, use ellipses. The Spanish menu mirrors this tree…

Key facts

Category
Audio & voice · Voiceovers
Output type
Audio
Level
Intermediate
Designed for
ChatGPT, Claude, Gemini and ElevenLabs
Verified in
Pending editorial verification
Variables
Company, Industry, Call reasons, Business hours, Languages and Brand tone
Language
English
Price
Free, no sign-up
Updated
Nov 7, 2026

Frequently asked questions

Why say the action before the key?

Callers hear what they need first and then hold on to the key to press. The other way round, they often forget the number before they know whether it applies to them.

Which audio format do phone systems need?

It depends on the provider: many telephony systems use 8 kHz mono WAV, while others accept MP3. Check your phone system’s documentation and convert the files if needed.

How many options should the menu have?

Ideally no more than four per level: with more, callers forget the first ones and end up pressing the option to speak to a person.

What is this prompt for?

Design your phone menu tree and get every prompt (welcome, options, after-hours, hold, invalid input) as a separate audio file, with numbers normalized for AI text to speech.

Which AI tools does it work with?

It is designed for ChatGPT, Claude, Gemini and ElevenLabs. We have not verified it in these tools yet, so results may vary between them.

How do I customize the prompt?

Click “Customize” and fill in the variables (Company, Industry, Call reasons, Business hours, Languages and Brand tone). The text updates instantly and you can then copy it in one click. Values are stored only in your browser.

Do I need experience to use it?

It is intermediate: it helps to know the topic so you can provide good context and review the result critically.

Is this prompt free?

Yes. Every HomeCasa Prompts prompt is free, with no sign-up or subscription. You can also use them in professional projects.

Can I trust the AI output?

Treat it as a high-quality draft: always check data, figures and important claims before publishing or making decisions.

Related prompts