DADS – Stutter Detection App

DADS (Detection and Analysis of Dysfluencies in Speech) is an AI-powered application designed to detect stuttering in speech recordings.
This app allows users to record or upload audio, analyze it, and visualize stuttering patterns efficiently.

📂 Project Structure

DADS/
│-- Scripts/             # Data Generation,Preprocessing,Utilities
│-- Model/               # Model training & Evaluation
|-- dataset/             # Extract the dataset here
│-- App/                 # Application Code (UI,API,Deployment)

SEP28k-dataset (from kaggle)

How to Use the App

Step 1: Clone the Repository

• Run the following command in your terminal (replace < repo-link > with your actual repository link):

git clone <repo-link>

---

Step 2: Create a Virtual Environment

• On macOS / Linux:

python3 -m venv .venv

• On Windows:

python -m venv .venv

---

Step 3: Activate the Virtual Environment and Install Dependencies

Activate the environment:

• macOS / Linux:

source .venv/bin/activate

• Windows:

.venv\Scripts\activate

Install the required packages:

pip install -r requirements.txt

---

Step 4: Run the App

Start the application by running:

• macOS / Linux

python3 App/run_app.py

• Windows

python App/run_app.py

---

Commit Guidelines

Follow this convention:

[TYPE] Commit message

Common types:

• [FIX] – Bug fixes
• [ADD] – New features
• [DOCS] – Documentation changes
• [MNT] – Code refactoring & Maintenance
• [TEST] – Tests additions or fixes