|
|
---
|
|
|
title: RAWI Voice to Story Generator
|
|
|
emoji: 📝
|
|
|
colorFrom: indigo
|
|
|
colorTo: green
|
|
|
sdk: gradio
|
|
|
sdk_version: 3.50.2
|
|
|
app_file: app.py
|
|
|
pinned: false
|
|
|
python_version: 3.9
|
|
|
---
|
|
|
|
|
|
# RAWI Voice to Story Generator
|
|
|
|
|
|
This Hugging Face Space converts Arabic voice recordings into polished stories using Whisper for speech recognition and DeepSeek API for creative text generation.
|
|
|
|
|
|
## How It Works
|
|
|
|
|
|
1. Upload or record an Arabic audio clip
|
|
|
2. The system transcribes the speech using OpenAI's Whisper model
|
|
|
3. The transcript is sent to DeepSeek API to generate an enhanced story
|
|
|
4. Both the original transcript and the generated story are displayed
|
|
|
|
|
|
## Setup
|
|
|
|
|
|
This Space requires a DeepSeek API key to work properly. When deploying:
|
|
|
|
|
|
1. Go to the Settings tab of your Space
|
|
|
2. Add your DeepSeek API key as a secret named `DEEPSEEK_API_KEY`
|
|
|
3. (Optional) If needed, change the API endpoint by adding `DEEPSEEK_API_URL`
|
|
|
|
|
|
## Deploying to Hugging Face Spaces
|
|
|
|
|
|
To deploy this application to Hugging Face Spaces:
|
|
|
|
|
|
1. Create a new Space on [Hugging Face](https://huggingface.co/spaces)
|
|
|
2. Select "Gradio" as the SDK
|
|
|
3. Upload the contents of this directory to your Space
|
|
|
4. Set the required secrets in the Space settings
|
|
|
5. Choose a suitable hardware tier (recommend at least CPU-M)
|
|
|
|
|
|
## Local Development
|
|
|
|
|
|
To run this project locally:
|
|
|
|
|
|
1. Clone this repository
|
|
|
2. Install dependencies: `pip install -r requirements.txt`
|
|
|
3. Set environment variables:
|
|
|
```
|
|
|
export DEEPSEEK_API_KEY=your_deepseek_api_key
|
|
|
```
|
|
|
4. Run the application: `python app.py`
|
|
|
|
|
|
## Technologies Used
|
|
|
|
|
|
- Whisper: AI-powered speech recognition model
|
|
|
- Gradio: Web interface for ML applications
|
|
|
- DeepSeek API: Arabic text generation and enhancement
|
|
|
|
|
|
## Note
|
|
|
|
|
|
This application is designed for Arabic language content. Using other languages may result in suboptimal performance.
|
|
|
|
