Article2Audio

Convert any article to audio with OpenAI's TTS API

Article2Audio is a web app that allows users to copy the URL to an article and receive back a lovely audio version of the article to download locally as a mp3 file. With some configuring and setting up an RSS feed, users can also automatically sync these audio files to their favorite podcast app as part of a personal podcast for easy listening on the go.

For ease of use, we also include a simple python script which allows users to directly convert article URLs to audio files without needing to spin up the web app.

It's currently very barebones and a work in progress, but the functionality is all there!

Simple web app (left), synced articles available as a personal podcast on Spotify, Podbean, etc. (right)

Sample Audio Output

The.Most.Precious.Resource.is.Agency.mp4

Installation: Web App

1. Clone the repository:

   git clone https://github.com/timf34/Article2Audio.git
   cd Article2Audio

2. Create .env files

   Client .env file_

   First create a .env file within the /client directory and add your localhost (or URL if you're deploying) to it:

    cd client
    echo "REACT_APP_API_URL=http://localhost:8001/api" > .env

   Server .env file

   Then create a .env file within the /server directory and add the following environment variables:

   cd server
   echo "OPENAI_KEY=your-openai-key" > .env
   echo "AWS_ACCESS_KEY_ID=your-aws-access-key" >> .env
   echo "AWS_SECRET_ACCESS_KEY=your-aws-secret-key" >> .env

   Note that the AWS keys are optional and for uploading to S3 for the RSS feed, but the OpenAI key is required for the app to function.

3. Setup Nginx (Optional)

   If you would like to set up a proper web server with a custom domain, you can use Nginx. We include scripts/setup_nginx.sh to help you set up Nginx on an Ubuntu server.

4. Build and run Docker

   This will spin up the web app:

   docker compose up --build

Installation: Configure RSS feed and sync to personal podcast

Assuming that you are running the web app on a server with a domain name, you can set up an RSS feed to sync the audio files to your favorite podcast app. We have built support for doing this using AWS S3, but you of course can switch this out for any other cloud storage provider, just ensure that your bucket is publicly accessible.

We will assume you know how to set up an S3 bucket and make it publicly accessible, and how to get your AWS keys. I'll include the S3 bucket policy here to help.

1. Update the `/server/data/rss.xml file with your own info

   Update the /server/data/rss.xml file with your own information (this will be obvious). This is the RSS feed that will be used to sync the audio files to your podcast app.

   To ensure that this has worked, visit www.yourdomain.com/rss.xml and check that the file is accessible.

2. Submit the RSS feed URL to podcast hosting services

   Submit the RSS feed URL to your favorite podcast hosting service (e.g. Spotify, Apple Podcasts, etc.) and you should be able to listen to your articles as a podcast!

   To do so with Spotify or Apple, you'll have to make an account. But many other podcast apps will allow you to add a custom RSS feed URL without needing an account, such as PodcastAddict.

The audio files will automatically upload to S3 once they've been generated, and should be available soon thereafter on your podcast app of choice.

Installation: Simple Python Script

We also include a simple python script that allows users to convert article URLs to audio files without needing to spin up the web app.

1. Clone the repository:

   git clone https://github.com/yourusername/Article2Audio.git
   cd Article2Audio

2. Add your OpenAI api key

   If you have not already, create a .env file in the /server directory and add your openAI key to it:

    cd server
    echo "OPENAI_KEY=your-openai-key" > .env

3. Install the required packages:

    pip install -r server/requirements.txt

Python Script Usage

The script convert_article_urls_to_audio.py allows for two modes of operation:

1. Default Mode: By default it will use the URLs specified in the HARDCODED_URLS list at the top of the file.

   python convert_article_urls_to_audio.py

2. Command Line Mode: Overrides hardcoded URLs by specifying URLs via the command line with the --urls option.

   python convert_article_urls_to_audio.py --urls https://www.thefitzwilliam.com/p/james-joyce-was-a-complicated-man "https://www.honest-broker.com/p/how-picasso-turned-me-into-a-strategy"

License

Article2Audio is licensed under a GPL-3.0 license.