msaelices/note-taker

Note taker

This Streamlit app allows users to upload an audio file, transcribe the audio using WhisperX or AssemblyAI, and summarize the transcription with GPT.

Installation

1. 1.Clone the repository:

git lfs install
git clone https://huggingface.co/spaces/msaelices/note-taker

1. 1.Change into the repository's directory:

cd note-taker

1. 1.Install the required packages:

pip install -r requirements.txt

1. 1.Set your needed secrets in the .env file. You can use the .env-default file as a reference:

cp .env-default .env

1. 1.Run the Streamlit app:

streamlit run app.py

The app will now be accessible at http://localhost:8501 in your browser.

Note

Once you have accessed to the web page, you can fill the required configuration (if not set in the .env file). The WhisperX engine need to install the pytorch package and will ask about some specific configuration, depending on your server, like the device (CPU or cuda), the precision (int8 vs float16), etc.