FiledimeGPT is made to work out of the box with Filedime, remotely or locally.

Step: Clone the github repo and navigate into it

git clone https://github.com/visnkmr/filegpt-filedime && cd filegpt-filedime

Step: Setup a Python Virtual Environment

Linux

python3 -m venv filedimegpt

Windows

py -m venv .venv 

Step: Activate the Python Virtual Environment

Linux

source ./filedimegpt/bin/activate

Windows

.\.venv\Scripts\activate   

Step: Install the Requirements

Linux

pip install -r requirements.txt

Windows

py -m pip install -r .\requirements.txt 

Step: Run this command

Linux

python app.py  

Windows

py .\app.py

If you only need filedimespeech then no need to install ollama which is described below,

Thanks to (https://github.com/imartinez/privateGPT), (https://github.com/jmorganca/ollama), (https://github.com/myshell-ai/MeloTTS) and (https://github.com/menloparklab/privateGPT-app)

Make sure to have Ollama running on your system from https://ollama.ai and by default it expects ollama to be running on port 11434

Step: Download/pull models (if you already have models loaded in Ollama, then not required)

ollama pull llama3

Modify the env file as neccesary by default it has the following values:

PERSIST_DIRECTORY=db
MODEL_TYPE=llama3
EMBEDDINGS_MODEL_NAME=all-MiniLM-L6-v2
MODEL_N_CTX=1000
SOURCE_DIRECTORY=source_documents
OLLAMA_URL=http://localhost:11434

The supported file extension that FileGPT can embed are:

• .csv: CSV,
• .docx: Word Document,
• .doc: Word Document,
• .enex: EverNote,
• .eml: Email,
• .epub: EPub,
• .html: HTML File,
• .md: Markdown,
• .msg: Outlook Message,
• .odt: Open Document Text,
• .pdf: Portable Document Format (PDF),
• .pptx : PowerPoint Document,
• .ppt : PowerPoint Document,
• .txt: Text file (UTF-8),