Skip to content

Latest commit

 

History

History
32 lines (22 loc) · 1.47 KB

File metadata and controls

32 lines (22 loc) · 1.47 KB

FluentFlow

FluentFlow

Overview

FluentFlow is a web application designed to assist users in enhancing their public speaking skills by evaluating their performance across various criteria such as body language, voice modulation, and speech content.

Features

  • Performance Evaluation:
    • Body Movement and Language: Analyzes gestures, posture, head gaze, and eye gaze to provide insights into non-verbal communication.
    • Voice Analysis: Measures aspects such as speed, vocal variation, pauses, and filler words to improve speech delivery.
    • Script and Language Evaluation: Examines the script's content, language variation, and overall coherence.

Technologies Used

Processing Libraries:

  • Text Processing: Utilized libraries such as NLTK, SpaCy, and XGBoost for linguistic analysis of the script.
  • Video Processing: Implemented MoviePy, PyTorch, SciPy, Gensim, and OpenCV to analyze body movements and gestures from video inputs.
  • Audio Processing: Used Librosa for in-depth vocal feature analysis.

Development Stack:

  • Frontend: Built the user interface using Flutter.
  • Backend: Developed the server-side logic with Django.

Developed by [Alya Almsouti - Hamza Almahrous - Osama Bazo - Yassin Abdulmahdi] in collaboration with Damascus University.

Demo

Hamza.Grad.Demo.-.Shows.3-1.mp4