# Speech

Turns your text into spoken audio files with OpenAI voices, for narration, voiceovers, phone menus and read-aloud clips.

- Kind: a skill for AI agents
- Version: 1.0, updated 2026-11-18
- Shared by: Workotron
- Written by: [OpenAI](https://github.com/openai/skills/tree/49f948faa9258a0c61caceaf225e179651397431/skills/.curated/speech)
- License: Apache-2.0
- Business area: Design and media
- Tags: #design #audio #codex
- Rank: not ranked yet, with 0 copies, 0 thumbs up and 0 thumbs down
- Permanent address: https://skills.workotron.com/speech

## Who is this for?

- Small businesses recording phone menu prompts or hold messages
- People making product demo voiceovers, explainer narration or accessible audio

## What it does

- Collects your exact text, the voice, the delivery style and the file format
- Writes short voice directions such as tone, pacing and emphasis without changing your text
- Runs a bundled script that calls the OpenAI Audio API for one clip or a batch
- Checks important clips for clarity, pacing and pronunciation and adjusts one thing at a time
- Saves the audio files and notes the text, directions and settings used

## When to use it

- When you need a voiceover or narration and do not want to record it yourself
- When you have many short lines, such as phone menu prompts, to turn into separate audio files
- Not for creating a custom voice: it uses only the built-in voices

## About this skill

It runs a Python script on your computer that sends your text to the OpenAI Audio API. It needs an OpenAI API key set on your computer and the openai Python package. Each request takes up to 4096 characters, so longer text is split.

The default voice is cedar, with marin for a brighter tone. The skill tells you to let listeners know the voice is AI-generated.

## Supported platforms

- Codex
- ChatGPT

## Install

Get https://skills.workotron.com/speech.zip and extract it into your skills folder. If you cannot unzip, get each file:

- https://skills.workotron.com/speech/files/LICENSE.txt
- https://skills.workotron.com/speech/files/SKILL.md
- https://skills.workotron.com/speech/files/agents/openai.yaml
- https://skills.workotron.com/speech/files/assets/speech-small.svg
- https://skills.workotron.com/speech/files/assets/speech.png
- https://skills.workotron.com/speech/files/references/accessibility.md
- https://skills.workotron.com/speech/files/references/audio-api.md
- https://skills.workotron.com/speech/files/references/cli.md
- https://skills.workotron.com/speech/files/references/codex-network.md
- https://skills.workotron.com/speech/files/references/ivr.md
- https://skills.workotron.com/speech/files/references/narration.md
- https://skills.workotron.com/speech/files/references/prompting.md
- https://skills.workotron.com/speech/files/references/sample-prompts.md
- https://skills.workotron.com/speech/files/references/voice-directions.md
- https://skills.workotron.com/speech/files/references/voiceover.md
- https://skills.workotron.com/speech/files/scripts/text_to_speech.py

## Discussion (0)
