Text to speech
Turns text into narration or voiceover with the OpenAI Audio API’s built-in voices, including batches for longer scripts.
When to use it
- You need narration for a video or demo.
- You want an audio version of written content.
- Several clips should share the same voice.
Creating custom voices is out of scope.
A practical example
You ask
“Record a calm voiceover for this 60-second product script.”
Your agent does
It picks a suitable voice, generates the audio and saves it ready to drop into your edit.
What gets installed
The whole folder is copied, so scripts, references and any license notices stay together.
- agents/openai.yaml299 bytes
- assets/speech-small.svg742 bytes
- assets/speech.png1,234 bytes
- LICENSE.txt10,776 bytes
- references/accessibility.md696 bytes
- references/audio-api.md902 bytes
- references/cli.md3,275 bytes
- references/codex-network.md1,148 bytes
- references/ivr.md708 bytes
- references/narration.md671 bytes
- references/prompting.md1,553 bytes
- references/sample-prompts.md991 bytes
- references/voice-directions.md2,155 bytes
- references/voiceover.md790 bytes
- scripts/text_to_speech.py15,592 bytes
- SKILL.md7,193 bytes
Before you start: Uses Python and needs an OPENAI_API_KEY for live calls.
Credit and license
Adapted from OpenAI’s published skills. It’s shared under the Apache License 2.0, and the license file is installed with it however you choose to install.