A selfhostable webapp which helps you read those uselessly long (by nature) voice messages with the power of AI.
Clone and check out the main branch and run docker build -t tl-dl .
to build the image.
Use the docker-compose.yml
as a starting point. You'll need an OPENAI API key to run it. Set it along your final URL, where the docker runs and set it as ORIGIN.
The app is exposed on PORT 3000. I recommend a reverse proxy like caddy or traefik in front for https and PORT handling.
Clone and check out the main branch. You'll need node-18 installed. node-20 will probably work too. Run npm ci
and npm run dev
to get started with a local environment. You should probably set your OPENAI_API_KEY as well OPENAI_API_KEY=YOURKEY npm run dev
.