Generate faceless videos on autopilot using Laravel.

If you're a more visual learner, you can check out this video where I go over the entire project:

Run the following commands to get started:

composer install
yarn install
cp .env.example .env
php artisan key:generate

Fill out the following .env variables:

REPLICATE_API_TOKEN= # https://replicate.com/account/api-tokens
OPENAI_API_KEY= # https://platform.openai.com/api-keys

Run the following command to start the dev services:

php artisan dev

This will start horizon, reverb and share the app using valet share. The app should be available over the internet for Replicate to access it.

In the faceless directory you will find a remotion project that is used to generate the videos. You can run the following command to start the Remotion studio:

cd faceless
yarn install
yarn start

By default, the faceless/src/example.json file gets loaded with a dummy story that you can use to test the video generation. If you want to generate a video with a different story, you can start the studio using the following parameter:

yarn start --props='{"json":"http:\/\/example.com\/story.json"}'

You can also control the FPS using the following parameter:

yarn start --props='{"json":"http:\/\/example.com\/story.json", "fps": 60}'

You may be required to obtain a Remotion license, refer to their license file for more info.

Remotion has a built-in Lambda renderer that can be used to render videos in the cloud.

Refer to the Remotion documentation for more information.

After you've set up a Lambda, you can add the following environment variable to your .env file:

REMOTION_APP_REGION=
REMOTION_APP_FUNCTION_NAME=
REMOTION_APP_SERVE_URL=

This way, RenderVideo.php will use the Lambda to render the video instead of the local Remotion instance.

• sabatinomasala/laravel-llm-prompt: This makes it easy to define LLM prompts.
• sabatinomasala/replicate-php: This is a PHP client for the Replicate API with 'run' functionality.
• sabatinomasala/dev-scripts-for-laravel: This makes it easy to run several services in parallel during development.