OpenBlue is a simple web interface developed in Flask to provide a graphical interface for interacting with the OpenAi API. It allows users to send requests to the OpenAi API and receive generated responses using GPT-3.5 and GPT-4 models.

OpenBlue is released under the MIT license. I do not take any responsibility for its usage.

• Text Chat using GPT-3.5 and GPT-4 models.
• Include file content in the chat conversation.
• Translate text.
• Translate DOCX, PDF, and TXT files.
• Translate only the text in DOCX without affecting the formatting.
• Transcribe audio and provide a summary or translation of it.

OpenBlue utilizes OpenAI APIs for its operation.

• Chat completions API for chat and translate.
• Trascription API for Audio trascription (using model Whisper)

OpenBlue requires an OpenAI API key. You need to store this key as an environment variable in your system with the name "API-KEY".

The creation of the API key is free through their official website. Each request incurs a variable cost (more information here).

Without ffmpeg, there might be some issues with transcribing audio and video files, but the rest would function identically.

First, ensure that Redis and ffmpeg are installed correctly .

• If you are using Arch, you can easily install Redis and Ffmpeg like this:

  sudo pacman -S redis
  sudo pacman -S ffmpeg

In order to run, you need Redis to be running.

redis-server

Then, download OpenBlue with this command and create a virtual environment:

git clone https://github.com/GiovanniMenon/OpenBlue.git
cd OpenBlue
python3.11 -m venv .venv

Activate Environment and install all the requirements in requirements.txt:

• Windows

  .venv\Scripts\activate
  pip install -r requirements.txt

• Linux/macOS

  . .venv/bin/activate
  pip install -r requirements.txt

Now you can start the application correctly.

python3.11 app.py

To create a user, you must have started the application at least once and then execute:

python3.11 db_config.py