Update README.md
Browse files
README.md
CHANGED
|
@@ -5,74 +5,10 @@ colorFrom: blue
|
|
| 5 |
colorTo: indigo
|
| 6 |
sdk: streamlit
|
| 7 |
sdk_version: 1.27.0
|
| 8 |
-
app_file:
|
| 9 |
pinned: false
|
| 10 |
---
|
| 11 |
|
| 12 |
# News Summarization and Text-to-Speech Application
|
| 13 |
|
| 14 |
-
This application extracts news articles about companies, performs sentiment analysis, conducts comparative analysis, and generates a text-to-speech output in Hindi.
|
| 15 |
-
|
| 16 |
-
## Features
|
| 17 |
-
|
| 18 |
-
- **News Extraction**: Scrapes the latest news articles about a company from reliable financial sources
|
| 19 |
-
- **Sentiment Analysis**: Analyzes the sentiment of each article (positive, negative, or neutral)
|
| 20 |
-
- **Topic Extraction**: Identifies key topics discussed in each article
|
| 21 |
-
- **Comparative Analysis**: Compares multiple articles to find differences in coverage and sentiment
|
| 22 |
-
- **Hindi Translation**: Translates the summary to Hindi
|
| 23 |
-
- **Text-to-Speech**: Generates audio output in Hindi
|
| 24 |
-
|
| 25 |
-
## How to Use
|
| 26 |
-
|
| 27 |
-
1. Enter a company name in the input field
|
| 28 |
-
2. Click "Analyze Company News"
|
| 29 |
-
3. View the sentiment analysis, article summaries, and comparative insights
|
| 30 |
-
4. Listen to the Hindi audio summary or read the translated text
|
| 31 |
-
|
| 32 |
-
## Technical Implementation
|
| 33 |
-
|
| 34 |
-
- **Backend**: FastAPI for API endpoints and data processing
|
| 35 |
-
- **Frontend**: Streamlit for user interface
|
| 36 |
-
- **NLP**: NLTK and Transformers for text analysis
|
| 37 |
-
- **Web Scraping**: BeautifulSoup4 for news extraction
|
| 38 |
-
- **Translation**: Deep Translator for Hindi translation
|
| 39 |
-
- **TTS**: gTTS (Google Text-to-Speech) with pyttsx3 fallback
|
| 40 |
-
|
| 41 |
-
## Local Development
|
| 42 |
-
|
| 43 |
-
### Prerequisites
|
| 44 |
-
|
| 45 |
-
- Python 3.9+
|
| 46 |
-
- Required libraries (install via `pip install -r requirements.txt`)
|
| 47 |
-
|
| 48 |
-
### Running Locally
|
| 49 |
-
|
| 50 |
-
1. Start the API server:
|
| 51 |
-
```
|
| 52 |
-
python api.py
|
| 53 |
-
```
|
| 54 |
-
|
| 55 |
-
2. Start the Streamlit app:
|
| 56 |
-
```
|
| 57 |
-
streamlit run app.py
|
| 58 |
-
```
|
| 59 |
-
|
| 60 |
-
3. Access the application at http://localhost:8501
|
| 61 |
-
|
| 62 |
-
## JSON Output Format
|
| 63 |
-
|
| 64 |
-
For developers interested in the API response format, you can get a standardized JSON output using:
|
| 65 |
-
|
| 66 |
-
```
|
| 67 |
-
python generate_json_output.py [COMPANY_NAME]
|
| 68 |
-
```
|
| 69 |
-
|
| 70 |
-
Or access the API endpoint:
|
| 71 |
-
```
|
| 72 |
-
POST /api/example_format
|
| 73 |
-
{"company_name": "COMPANY_NAME"}
|
| 74 |
-
```
|
| 75 |
-
|
| 76 |
-
## License
|
| 77 |
-
|
| 78 |
-
MIT License
|
|
|
|
| 5 |
colorTo: indigo
|
| 6 |
sdk: streamlit
|
| 7 |
sdk_version: 1.27.0
|
| 8 |
+
app_file: app_spaces.py
|
| 9 |
pinned: false
|
| 10 |
---
|
| 11 |
|
| 12 |
# News Summarization and Text-to-Speech Application
|
| 13 |
|
| 14 |
+
This application extracts news articles about companies, performs sentiment analysis, conducts comparative analysis, and generates a text-to-speech output in Hindi.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|