README.md

metadata

title: AI-powered ASL text-to-video Generator
emoji: 🐻
colorFrom: blue
colorTo: yellow
sdk: gradio
sdk_version: 5.34.2
app_file: app.py
pinned: false
license: apache-2.0

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

AI-SL API

Convert natural language English into American Sign Language (ASL) videos using AI!

View our full repo for the AI-SL Project created for the Berkeley AI Hackathon 2025 🚀 here: AI-SL Repo

Features

Dual Input Support with Optional File Upload

The app accepts both text input and file uploads with flexible options:

• Text Input: Type or paste text directly into the interface (always available)
• File Upload: Upload documents (PDF, TXT, DOCX, EPUB)

Video Output Options

The Gradio interface provides multiple ways for users to receive and download the generated ASL videos:

1. R2 Cloud Storage

• Videos are automatically uploaded to Cloudflare R2 storage
• Returns a public URL that users can download directly
• Videos persist and can be shared via URL
• Includes a styled download button in the interface

2. Base64 Encoding (Alternative)

• Videos are embedded as base64 data directly in the response
• No external storage required
• Good for smaller videos or when you want to avoid cloud storage
• Can be downloaded directly from the interface

3. Programmatic Access

Users can access the video output programmatically using:

from gradio_client import Client

# Connect to the running interface
client = Client("http://localhost:7860")

# Upload a document and get results
result = client.predict(
    "path/to/document.pdf",
    api_name="/predict"
)

# The result contains: (json_data, video_output)
json_data, video_url = result

# Download the video
import requests
response = requests.get(video_url)
with open("asl_video.mp4", "wb") as f:
    f.write(response.content)

Example Usage

Web Interface

1. Visit your Space URL
2. Choose input method:
   • Text: Type or paste text in the text box (always available)
   • File: Check "Enable file upload" and upload a document (optional)
3. Click "Submit"
4. Download the resulting video

Programmatic Access with Optional File Upload

from gradio_client import Client

# Connect to your hosted app
from gradio_client import Client, handle_file
client = Client("deenasun/ai-sl-api")

# Text input only (file upload disabled)
result = client.predict(
    text="Hello world! This is a test.",  # Text input
    file=None,                            # File input (None since disabled)
    api_name="/predict"
)

# File input only (file upload enabled)
result = client.predict(
    text="",                              # Text input (empty)
    file=handle_file("document.pdf"),     # File input
    api_name="/predict"
)

# Both inputs (text takes priority)
result = client.predict(
    "Quick text",                    # Text input
    "document.pdf",                  # File input
    api_name="/predict"
)

See example_usage.py and example_usage_dual_input.py for complete examples of how to:

• Download videos from URLs
• Process base64 video data
• Use the interface programmatically
• Perform further video processing
• Handle both text and file inputs
• Use optional file upload functionality

Requirements

• Python 3.7+
• Required packages listed in requirements.txt
• Cloudflare R2 credentials (for cloud storage option)
• Supabase credentials for video database

Setup

1. Install dependencies: pip install -r requirements.txt
2. Set up environment variables in .env file
3. Run the interface: python app.py

Video Processing

Once you have the video file, you can:

• Upload to YouTube, Google Drive, or other services
• Analyze with OpenCV for computer vision tasks
• Convert to different formats
• Extract frames for further processing
• Add subtitles or overlays