프로메테우스 AI 발표연습 어플리케이션 개발

• General info
• Technologies
• Setup
• Members

1. Clone the repository :

git clone https://github.com/use08174/SpeechPT.git

2. Install dependencies:

pip install -r requirements.txt

3. Start the development server:

cd SpeechPT/backend/app
uvicorn main:app --reload

Our project has two key features: presentation analysis and text summarization.

After STT using Google STT API, we implemented three functions as follow.

• Speed : Calculate cpm to analyze speed of the speech. (cpm : character per minute)
• Pause : Calculate number of pauses and duration of pauses during speech.
• Filler words : Detect korean filler words in the speech.

• Face Expression : Analyze face expression of the presenter.
• Gaze Tracking : Tracking gaze of the presenter.

• Summarize text
  1. Extract text from a document (pdf,word) and summarize text by paragraph using Kobart
  2. Check the spelling and save the final summary to the document

This is our notion page for our project archive. : Notion

More about Prometheus : https://prometheus-ai.net/