Amruthayenikonda/News-Summarization-Hindi-TTS

📰 News Sentiment Analysis with Hindi TTS 🎙️

📌 Project Overview

This project is a news sentiment analysis tool that scrapes recent news articles for a given company, summarizes them, performs sentiment analysis, and generates a Hindi TTS (Text-to-Speech) summary. The analysis helps users understand media coverage trends and sentiment distribution.

🔥 Features

• —Scrape news articles from selected sources.

• —Summarize articles using the BART model.

• —Perform sentiment analysis using TextBlob.

• —Generate Hindi speech with gTTS.

• —Provide a comparative analysis of different articles.

• —FastAPI backend with endpoints for data access.

• —Deployed on Hugging Face Spaces for easy access.

🚀 Tech Stack

• —Python (FastAPI, BeautifulSoup, Requests, Pandas, NLTK)

• —YAKE (Keyword extraction)

• —TextBlob (Sentiment analysis)

• —Facebook BART Model (Summarization)

• —gTTS (Hindi Text-to-Speech)

• —FastAPI (Backend API development)

• —Streamlit (User interface)

• —Hugging Face Spaces (Deployment)

📥 Installation & Setup

🛠️ 1. Clone the Repository

📦 2. Install Dependencies

▶️ 3. Run the FastAPI Server

API will be available at: http://127.0.0.1:8000/docs

🎭 4. Run the Streamlit UI (Optional)

run.py is there to serve as the entry point for running the application, handling initialization, and managing API or UI execution. 🚀

🌍 How to Use

1️⃣ Access the API

Visit Your Hugging Face Space to test it directly.

2️⃣ API Endpoints

Method

Endpoint

Description

GET

/scrape_news/{company}

Scrape news articles for a company

GET

/get_news/{company}

Get sentiment analysis report

GET

/get_tts/{company}

Get Hindi speech summary

3️⃣ Example Request

📊 Model Details

🔹 Summarization Model

• —Uses Facebook BART (facebook/bart-large-cnn) for text summarization.

• —Extracts key information from scraped news articles.

🔹 Sentiment Analysis

• —Uses TextBlob for polarity-based sentiment scoring.

• —Classifies articles as Positive, Negative, or Neutral.

🔹 Hindi TTS (Text-to-Speech)

• —Uses gTTS (Google Text-to-Speech) to generate Hindi audio.

• —Creates a spoken summary of sentiment analysis.

⚠️ Assumptions & Limitations

• —Scraped articles depend on website structure and may fail if changes occur.

• —Sentiment analysis is based on text polarity and may not always capture nuances.

• —Only supports English text input but generates Hindi speech.

🚀 Deployment

🔗 Live Demo You can try the application live on Hugging Face Spaces: 🖥️ News Summarization Hindi TTS

No account needed to access API docs and test endpoints!

💡 Feel free to contribute or report issues! 🎯