Rule: Elevenlabs Dialogue Creator (Elevenlabs Field)

Base data:

Summary:
The Elevenlabs Dialogue Creator takes a prompt, that can include some context and generates a script for the Elevenlabs field that is then creating the audio as spoken word using Elevenlabs.

Module needed:
AI Interpolator Elevenlabs Field, a submodule to ElevenLabs Field

Field types to populate:
Elevenlabs (From ElevenLabs Field module)

Base Fields types to use as context:

• File

Extra Requirements:
You will need AI Interpolator OpenAI to be installed. You need both an OpenAI account and an Elevenlabs account.

Prompting:

The prompt in this case requires some things to be stated very clearly - one is that the voices needs to be give in data source form. These can be found under the placeholders fieldset.

Be sure to mention how long they should speak and how many dialogues back an forth. Name the voices and mention who says what.

If its for audio summarizations, you can just name the voice and ask it to summarize in the way you want.

If you use Advanced Prompting, not that the AI Interpolator Elevenlabs Field module comes with a new field type that is all the ElevenLabs voices, meaning that you can give the editorial user a choice to choose this from one/multiple drop downs, which voices to use.

Extra Settings:

OpenAI Model

The OpenAI Model to use. Even if this is a choice, a GPT-4 model should always be used to generate elevenlabs dialogues.

OpenAI Role

The OpenAI Role can be filled out, if you want some explanation of the agent generating data for you.

Extra Advanced Settings:

None

Possible example use cases:

• Create a podcast from a webpage, by scraping it and using it as context.
• Create an audio summarization of a PDF or Word file using the AI Interpolator ConvertAPI.