Automated AI Voiceover Video Creation with Creatomate and ElevenLabs

Description and use

A guide to automating AI voiceover video creation using Creatomate and ElevenLabs. Enables generation of realistic voices and synchronized subtitles via a single API request.

Problem

Traditional text-to-speech APIs often produce robotic, flat audio that lacks human emotion and depth, making them unsuitable for engaging social media content.

Outcome

A fully automated workflow that generates high-quality vertical videos with realistic AI voiceovers and synchronized animated captions using a single API call.

Workflow

• Sign up for ElevenLabs and select or clone a voice, then copy the Voice ID.
• Generate an ElevenLabs API key with Text-to-Speech access enabled.
• Log in to Creatomate and connect the ElevenLabs account via Project Settings using the API key.
• Create a new video template in Creatomate, selecting a 'Short-Form Voice Over' template with 9:16 vertical aspect ratio.
• Configure voiceover elements in the template to use the ElevenLabs provider and the specific custom Voice ID.
• Set subtitle elements to source text from the corresponding voiceover elements.
• Configure image elements as dynamic placeholders.
• Send a POST request to the Creatomate API with the template ID and modifications (text for voiceovers, URLs for images).
• Creatomate processes the request, calls ElevenLabs for audio, generates subtitles, and renders the final video.

Evidence

Creatomate uses ElevenLabs to generate the voice, create the subtitles, and put it all together in a video – all with just a single API call.

Source

ElevenLabs' text-to-speech AI is capable of generating speech that is rich in emotion, tone, and depth.

Source