Labeled IVR audio
Give every branch a spoken prompt
IVR Voice Pack turns your company name and menu branches into a labeled set of phone-tree prompts in one brand voice, about twelve by default, covering welcome, main menu, hold, transfer, after-hours, error, and goodbye. Each prompt arrives as its own MP3 named for its menu position, in a library voice or a voice clone of your own from a 10-second to 5-minute sample. IVR Voice Pack works in Claude Code, Codex, Cursor, and other agents that run Skills.
Install IVR Voice Pack and tell me when I need to sign in: https://beatra.ai/skills/ivr-voice-pack/install.md
For agents:Install guideMarkdown version
Send it to WorkBuddy, Codex, Claude Code, or another agent to install. Sign in once the first time you use it.GitHub ↗Docs ↗
A prompt pack the switch can play
Cover the whole tree
Welcome, menu, hold, transfer, after-hours, and error sit in one labeled pack.
Keep one brand voice
Callers hear the same voice from the greeting through the goodbye.
Label every prompt
Name files by slot so the switch can take each clip.
How it is made
Write every branch the caller will hear
Turn the company name and menu keys into a list of welcome, menu, hold, transfer, after-hours, and error prompts.
Name the line the caller hears · List the menu keys · Keep after-hours and error lines
Choose one voice the tree can keep
Cast a brand voice, lock name and extension readings, and hear that voice on every prompt.
Choose one brand voice · Lock names and extensions · Hear the welcome first
Hand over clips the switch can place
Deliver numbered audio with slot IDs, ready for the phone tree.
Name files by slot · Keep one voice through the pack · Hear names and keys stay clear
Hotlines, auto attendants, call-center trees
Hotline menus
Give sales, support, and hours their own spoken keys.
Auto attendants
Record a default twelve-prompt tree for a new line.
Call-center trees
Refresh hold, transfer, and after-hours lines when the hours change.
Inputs and outputs
Input · Text — Company or line name, menu branches, and readings for names and terms
Input · Audio — Voice sample for a cloned voice (optional)
- Specifications
- 10–300 s; up to 20 MB each
Output · Audio — Labeled phone-tree prompts in one brand voice, one file per prompt
- Formats
- MP3
| Input or output | Type | Formats | Specifications |
|---|---|---|---|
| Input | Text — Company or line name, menu branches, and readings for names and terms | — | — |
| Input | Audio — Voice sample for a cloned voice (optional) | — | 10–300 s; up to 20 MB each |
| Output | Audio — Labeled phone-tree prompts in one brand voice, one file per prompt | MP3 | — |
Specifications last checked · Sources: Model catalog · Documentation · Terms of Service
Questions
What do I need to start?
The company or line name, the menu branches (or a default twelve-prompt tree), the spoken language, and any names that must be read the same way.
Which voice reads the prompts?
One brand voice reads every prompt, from the welcome through hold, transfer, after-hours, and error: a library voice or a clone of your own.
What will I receive?
About twelve labeled prompts covering welcome, menu, hold, transfer, after-hours, and error, ready to drop into the switch.
Related Skills
