Initial commit: media-gen-api with Streamlit UI

media-gen-api/.devcontainer/devcontainer.json

@@ -0,0 +1,33 @@
+{
+  "name": "Python 3",
+  // Or use a Dockerfile or Docker Compose file. More info: https://containers.dev/guide/dockerfile
+  "image": "mcr.microsoft.com/devcontainers/python:1-3.11-bullseye",
+  "customizations": {
+    "codespaces": {
+      "openFiles": [
+        "README.md",
+        "app.py"
+      ]
+    },
+    "vscode": {
+      "settings": {},
+      "extensions": [
+        "ms-python.python",
+        "ms-python.vscode-pylance"
+      ]
+    }
+  },
+  "updateContentCommand": "[ -f packages.txt ] && sudo apt update && sudo apt upgrade -y && sudo xargs apt install -y <packages.txt; [ -f requirements.txt ] && pip3 install --user -r requirements.txt; pip3 install --user streamlit; echo '✅ Packages installed and Requirements met'",
+  "postAttachCommand": {
+    "server": "streamlit run app.py --server.enableCORS false --server.enableXsrfProtection false"
+  },
+  "portsAttributes": {
+    "8501": {
+      "label": "Application",
+      "onAutoForward": "openPreview"
+    }
+  },
+  "forwardPorts": [
+    8501
+  ]
+}

media-gen-api/.gitignore

@@ -0,0 +1,44 @@
+# Python
+*.pyc
+__pycache__/
+*.pyo
+*.pyd
+*.log
+
+# Virtual env
+.venv/
+venv/
+ENV/
+env/
+
+# Environment secrets
+.env
+
+# Streamlit cache
+.streamlit/
+.metadata/
+
+# Pytest cache
+.pytest_cache/
+
+# VS Code / PyCharm / IDE files
+.vscode/
+.idea/
+.project
+.pydevproject
+
+# RStudio / Mac / Windows artifacts
+.Rhistory
+.DS_Store
+Thumbs.db
+
+# Output folders
+outputs/
+
+
+logs/
+__pycache__/
+outputs/
+generated/
+*.pyc
+*.log

media-gen-api/.vs/VSWorkspaceState.json

@@ -0,0 +1,6 @@
+{
+  "ExpandedNodes": [
+    ""
+  ],
+  "PreviewInSolutionExplorer": false
+}

media-gen-api/.vs/ai_gov_comm/v17/DocumentLayout.json

@@ -0,0 +1,12 @@
+{
+  "Version": 1,
+  "WorkspaceRootPath": "C:\\Users\\fimba\\OneDrive\\Desktop\\PythonProjects\\ai_gov_comm\\",
+  "Documents": [],
+  "DocumentGroupContainers": [
+    {
+      "Orientation": 0,
+      "VerticalTabListWidth": 256,
+      "DocumentGroups": []
+    }
+  ]
+}

media-gen-api/Dockerfile

@@ -0,0 +1,34 @@
+# Use stable Python image
+FROM python:3.10
+
+# Install OS dependencies for Pillow and others
+RUN apt-get update && apt-get install -y \
+    build-essential \
+    libjpeg-dev \
+    zlib1g-dev \
+    libpng-dev \
+    libfreetype6-dev \
+    && rm -rf /var/lib/apt/lists/*
+
+# Set working directory
+WORKDIR /app
+
+# Copy requirement files first for caching
+COPY requirements.txt ./
+
+# Install Python dependencies
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Copy the app files
+COPY . /app
+
+# Expose port (the port is dynamic and set by the platform, but 8000 is a good fallback)
+EXPOSE 8000
+
+# Create a script to start Streamlit with dynamic PORT
+RUN echo '#!/bin/bash\n\
+export PORT=${PORT:-8000}\n\
+streamlit run app.py --server.port=$PORT --server.address=0.0.0.0' > /start.sh && chmod +x /start.sh
+
+# Run the script
+CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "8000"]

media-gen-api/PRIVACY_POLICY.md

@@ -0,0 +1,38 @@
+
+---
+
+### ✅ `PRIVACY_POLICY.md`
+
+```markdown
+# 🔐 Privacy Policy for OSNarayana Media Generator
+
+Effective Date: July 19, 2025
+
+Your privacy is important to us. This app is built to prioritize user data security and transparency.
+
+## 1. What We Collect
+
+- **No personal data** is collected, stored, or shared by this application.
+- All prompts, audio, and media are **processed locally** or via **user-provided API keys** to external services (e.g., ElevenLabs, Unsplash).
+
+## 2. API Usage
+
+- If you use external services (like ElevenLabs or Unsplash), you are subject to their respective [Terms of Service](https://www.elevenlabs.io/terms) and [Privacy Policies](https://www.elevenlabs.io/privacy).
+- Your API keys are stored **locally** in your environment file (`.env`) and never uploaded.
+
+## 3. Data Storage
+
+- Generated images, audio, and videos are stored **only on your local machine** under the `outputs/` directory.
+- You can delete any generated content at your discretion.
+
+## 4. Analytics & Ads
+
+- This app contains **no ads**, **no tracking**, and **no analytics**.
+
+## 5. Changes to Policy
+
+Any updates to this policy will be reflected in this file on the [GitHub repository](https://github.com/your-username/osnarayana-media-generator).
+
+## 6. Contact
+
+For questions or issues, contact: **[email protected]**

media-gen-api/Procfile

@@ -0,0 +1 @@
+web: streamlit run app.py --server.port=8000 --server.address=0.0.0.0

media-gen-api/README.md

@@ -0,0 +1,94 @@
+# 🎙️ Media Generation API
+
+A FastAPI-based backend to generate audio, images, video, and PPT from user inputs. 
+Supports BLEU/CLIP metrics, token-based authentication, and stores metadata in SQLite/Postgres.
+
+A modular, RESTful FastAPI solution that converts text input into:
+- 🎥 Video
+- 🖼️ Image/Graphics
+- 🔊 Audio
+
+
+---
+
+## 🚀 Features
+
+- Text → Video: Tone, domain, and environment-aware video generation.
+- Text → Audio: Context-aware voice synthesis with emotional tone and language support.
+- Text → Graphics: Visual generation using parameter-based prompts.
+- BLEU/CLIP metrics for prompt-output fidelity.
+- Token-based authentication for secure API use.
+- Dockerized for easy deployment
+- Optional Streamlit/React UI
+- Swagger UI: `http://localhost:8000/docs`
+
+---
+
+### 📁 Project Structure
+media-gen-api/
+├── app/
+│   ├── api/v1/               # Versioned API endpoints
+│   ├── auth/                 # Token-based auth
+│   ├── services/             # Core media generation logic
+│   └── main.py               # FastAPI entry point
+├── tests/                    # Unit/integration tests
+├── requirements.txt
+└── README.md
+
+---
+
+## 📦 Installation
+🚀 Run Locally
+1. Clone repo & create virtual environment
+
+git clone https://github.com/yourorg/media-gen-api.git
+cd media-gen-api
+python -m venv .venv
+source .venv/bin/activate  # On Windows: .venv\Scripts\activate
+
+2. Install dependencies
+
+pip install -r requirements.txt
+
+3. Run the API
+
+uvicorn app.main:app --reload
+
+Access docs: http://127.0.0.1:8000/docs
+
+---
+### 🔐 Authentication
+Use Bearer <your_token> in the Authorize button or headers.
+
+---
+### 📡 API Endpoints Summary
+| Endpoint                  | Method | Description               |
+|--------------------------|--------|---------------------------|
+| /api/v1/audio/generate   | POST   | Generate audio from text |
+| /api/v1/image/generate   | POST   | Generate image from text |
+| /api/v1/video/generate   | POST   | Generate video from text |
+| /api/v1/download         | GET    | Download generated file  |
+
+---
+###📦 Deployment (Streamlit/Optional UI)
+Option 1: Run with Streamlit (for demo)
+streamlit run streamlit_ui.py
+
+Option 2: Docker (Production-ready)
+docker build -t media-gen-api .
+docker run -p 8000:8000 media-gen-api
+
+---
+### 📊 Metrics Logging (Optional)
+- BLEU score and CLIPScore (WIP)
+- Latency, GPU/CPU tracking
+- Log file: logs/generation.log
+
+---
+#### 📋 Submission Checklist
+- ✅ RESTful modular architecture
+- ✅ Multi-format (MP4, PNG, WAV)
+- ✅ Token Auth + Swagger UI
+- ✅ Compatible with DD/PIB via API
+- ✅ Streamlit demo app (optional)
+

media-gen-api/app/api/v1/audio.py

@@ -0,0 +1,27 @@
+# app/api/v1/audio.py
+from fastapi import APIRouter, HTTPException, Query, Depends, Request
+from pydantic import BaseModel
+from app.services.audio_service import generate_audio_file  # ✅ Only this import
+from app.auth.auth import verify_token
+
+
+router = APIRouter()
+
+class AudioInput(BaseModel):
+    text: str
+    voice: str = "default"
+    language: str = "en"
+
+@router.post("/generate", dependencies=[Depends(verify_token)])
+def generate_audio_endpoint(payload: AudioInput):
+    try:
+        file_path = generate_audio_file(payload.text, payload.voice, payload.language)
+
+        return {
+            "status": "success",
+            "type": "audio",
+            "file_path": file_path,
+            "download_url": f"/api/v1/download?file_path={file_path}"
+        }
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))

media-gen-api/app/api/v1/download.py

@@ -0,0 +1,23 @@
+from fastapi import APIRouter, HTTPException, Query, Depends, Request
+from fastapi.responses import FileResponse
+from app.auth.auth import verify_token
+import os
+
+router = APIRouter()
+
+@router.get("/", dependencies=[Depends(verify_token)])
+def download_file(file_path: str = Query(..., description="Relative path from project root")):
+    # Sanitize the input path
+    file_path = os.path.normpath(file_path)
+
+    # Absolute path (from project root)
+    abs_path = os.path.join(os.getcwd(), file_path)
+
+    if not os.path.isfile(abs_path):
+        raise HTTPException(status_code=404, detail="File not found")
+
+    return FileResponse(
+        path=abs_path,
+        filename=os.path.basename(abs_path),
+        media_type='application/octet-stream'
+    )

media-gen-api/app/api/v1/image.py

@@ -0,0 +1,19 @@
+from fastapi import APIRouter, HTTPException, Query, Depends, Request
+from pydantic import BaseModel
+from app.services.image_service import generate_image_file
+from app.auth.auth import verify_token
+
+router = APIRouter()
+
+class ImageInput(BaseModel):
+    prompt: str
+    style: str = "default"
+
+@router.post("/generate", dependencies=[Depends(verify_token)])  # ✅ FIX: POST, not GET
+def generate_image(payload: ImageInput):
+    filename = generate_image_file(payload.prompt, payload.style)
+    return {
+        "message": "Image generated successfully",
+        "filename": filename,
+        "download_url": f"/api/v1/download?file_path=generated/image/{filename}"
+    }

media-gen-api/app/api/v1/metrics.py

@@ -0,0 +1,24 @@
+# app/api/v1/metrics.py
+from fastapi import APIRouter, HTTPException, Query, Depends, Request
+from typing import List
+from sklearn.metrics import accuracy_score
+from nltk.translate.bleu_score import sentence_bleu
+from sentence_transformers import SentenceTransformer, util
+from app.auth.auth import verify_token
+
+router = APIRouter()
+model = SentenceTransformer("clip-ViT-B-32")  # for CLIP-like semantic score
+
+@router.post("/evaluate/bleu", dependencies=[Depends(verify_token)])
+def compute_bleu(reference: str, candidate: str):
+    ref_tokens = [reference.split()]
+    cand_tokens = candidate.split()
+    score = sentence_bleu(ref_tokens, cand_tokens)
+    return {"metric": "BLEU", "score": score}
+
+@router.post("/evaluate/clipscore")
+def compute_clip_score(reference: str, candidate: str):
+    ref_emb = model.encode(reference, convert_to_tensor=True)
+    cand_emb = model.encode(candidate, convert_to_tensor=True)
+    score = util.cos_sim(ref_emb, cand_emb).item()
+    return {"metric": "CLIPScore", "score": score}

media-gen-api/app/api/v1/ppt.py

@@ -0,0 +1,24 @@
+# app/api/v1/ppt.py
+from fastapi import APIRouter, HTTPException, Query, Depends, Request
+from pydantic import BaseModel
+from typing import List
+from app.services.ppt_service import generate_ppt_file
+from app.auth.auth import verify_token
+
+router = APIRouter()
+
+class Slide(BaseModel):
+    title: str
+    content: str
+
+class PPTInput(BaseModel):
+    slides: List[Slide]
+
+@router.post("/generate", dependencies=[Depends(verify_token)])
+def generate_ppt(payload: PPTInput):
+    filename = generate_ppt_file([slide.dict() for slide in payload.slides])
+    return {
+        "message": "PPT generated successfully",
+        "filename": filename,
+        "download_url": f"/api/v1/download?file_path=generated/ppt/{filename}"
+    }

media-gen-api/app/api/v1/utils.py

@@ -0,0 +1,9 @@
+# app/api/v1/utils.py
+from fastapi.responses import FileResponse
+
+def download_file(file_path: str):
+    return FileResponse(
+        path=file_path,
+        filename=file_path.split("/")[-1],
+        media_type="application/octet-stream"
+    )

media-gen-api/app/api/v1/video.py

@@ -0,0 +1,20 @@
+from fastapi import APIRouter, HTTPException, Query, Depends, Request
+from pydantic import BaseModel
+from app.services.video_service import generate_video_file
+from app.auth.auth import verify_token
+
+
+router = APIRouter()
+
+class VideoInput(BaseModel):
+    script: str
+    duration: int = 10
+
+@router.post("/generate", dependencies=[Depends(verify_token)])  # ✅ FIX: POST, not GET
+def generate_video(payload: VideoInput):
+    filename = generate_video_file(payload.script, payload.duration)
+    return {
+        "message": "Video generated successfully",
+        "filename": filename,
+        "download_url": f"/api/v1/download?file_path=generated/video/{filename}"
+    }

media-gen-api/app/auth/auth.py

@@ -0,0 +1,32 @@
+from fastapi import Request, HTTPException, status
+from fastapi import Depends, HTTPException, status
+from fastapi.security import HTTPBearer, HTTPAuthorizationCredentials
+
+
+# Replace with a more secure method later (e.g., from .env)
+API_TOKEN = "my_secure_token_123"  # we can generate a random one using uuid or secrets
+
+def verify_token(request: Request):
+    token = request.headers.get("Authorization")
+    if not token or token.replace("Bearer ", "") != API_TOKEN:
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail="Invalid or missing token",
+        )
+    return request
+
+security = HTTPBearer()
+
+VALID_TOKENS = {"token123", "mysecuretoken"}  # Your valid tokens
+
+def auth_required(
+    credentials: HTTPAuthorizationCredentials = Depends(security),
+):
+    token = credentials.credentials
+    if token not in VALID_TOKENS:
+        raise HTTPException(
+            status_code=status.HTTP_401_UNAUTHORIZED,
+            detail="Invalid or missing token"
+        )
+    return token
+

media-gen-api/app/core/config.py

@@ -0,0 +1,6 @@
+import os
+
+class Settings:
+    API_KEY: str = os.getenv("API_KEY", "dummy-api-key")
+
+settings = Settings()

media-gen-api/app/db.py

@@ -0,0 +1,7 @@
+from sqlalchemy import create_engine
+from sqlalchemy.orm import sessionmaker
+
+DATABASE_URL = "sqlite:///./media_gen.db"
+
+engine = create_engine(DATABASE_URL, connect_args={"check_same_thread": False})
+SessionLocal = sessionmaker(bind=engine, autoflush=False)

media-gen-api/app/main.py

@@ -0,0 +1,70 @@
+# app/main.py
+from fastapi import FastAPI
+from app.api.v1 import audio, video, image, ppt, metrics, download
+import logging
+import os
+import sentry_sdk
+from fastapi import FastAPI, Depends
+from fastapi.security import HTTPBearer
+from app.api.v1.audio import router as audio_router
+from app.api.v1.video import router as video_router
+from app.api.v1.image import router as image_router
+from app.api.v1.ppt import router as ppt_router
+from app.api.v1.metrics import router as metrics_router
+from app.api.v1.download import router as download_router
+
+from fastapi.openapi.utils import get_openapi
+from fastapi.security import HTTPBearer
+
+
+
+# Create logs directory if it doesn't exist
+os.makedirs("logs", exist_ok=True)
+
+app = FastAPI(
+    title="Media Generator API",
+    description="Generate audio, video, image, and PPT from text",
+    version="1.0.0"
+)
+
+app.include_router(audio_router, prefix="/api/v1/audio", tags=["Audio"])
+app.include_router(video_router, prefix="/api/v1/video", tags=["Video"])
+app.include_router(image_router, prefix="/api/v1/image", tags=["Image"])
+app.include_router(ppt_router, prefix="/api/v1/ppt", tags=["PPT"])
+app.include_router(metrics_router, prefix="/api/v1/metrics", tags=["Metrics"])
+app.include_router(download_router, prefix="/api/v1/download", tags=["Download"])
+
+logging.basicConfig(
+    level=logging.INFO,
+    format="%(asctime)s [%(levelname)s] %(message)s",
+    handlers=[logging.FileHandler("logs/app.log"), logging.StreamHandler()]
+)
+
+sentry_sdk.init(dsn="https://a1d84892719d25dca7c04804931d2e82@o4509752680775680.ingest.de.sentry.io/4509752685166672", traces_sample_rate=1.0)
+
+
+security = HTTPBearer()
+
+
+def custom_openapi():
+    if app.openapi_schema:
+        return app.openapi_schema
+    openapi_schema = get_openapi(
+        title="Media Generator API",
+        version="1.0.0",
+        description="Generate audio, video, image, and PPT from text",
+        routes=app.routes,
+    )
+    openapi_schema["components"]["securitySchemes"] = {
+        "HTTPBearer": {
+            "type": "http",
+            "scheme": "bearer"
+        }
+    }
+    for path in openapi_schema["paths"].values():
+        for method in path.values():
+            method["security"] = [{"HTTPBearer": []}]
+    app.openapi_schema = openapi_schema
+    return app.openapi_schema
+
+app.openapi = custom_openapi

media-gen-api/app/models.py

@@ -0,0 +1,14 @@
+from sqlalchemy import Column, Integer, String, DateTime
+from sqlalchemy.ext.declarative import declarative_base
+from datetime import datetime
+
+Base = declarative_base()
+
+class MediaGeneration(Base):
+    __tablename__ = "media_generations"
+
+    id = Column(Integer, primary_key=True, index=True)
+    media_type = Column(String)
+    prompt = Column(String)
+    file_path = Column(String)
+    timestamp = Column(DateTime, default=datetime.utcnow)

media-gen-api/app/services/audio_service.py

@@ -0,0 +1,38 @@
+# app/services/audio_service.py
+from gtts import gTTS
+import os
+from datetime import datetime
+from app.db import SessionLocal
+from app.models import MediaGeneration
+import logging
+logger = logging.getLogger(__name__)
+
+def generate_audio_file(text: str, voice: str = "default", language: str = "en") -> str:
+    try:
+        tts = gTTS(text=text, lang=language, slow=False)
+        timestamp = datetime.now().strftime("%Y%m%d%H%M%S")
+        filename = f"audio_{timestamp}.mp3"
+        output_dir = "generated/audio"
+        os.makedirs(output_dir, exist_ok=True)
+        file_path = os.path.join(output_dir, filename)
+        tts.save(file_path)
+        logger.info(f"Generated Audio: {filename}")
+        return file_path
+    except:
+        logger.error(f"Audio Generation Failed: {str(e)}")
+        raise
+
+
+from app.db import SessionLocal
+from app.models import MediaGeneration
+
+def save_metadata(media_type, prompt, file_path):
+    db = SessionLocal()
+    record = MediaGeneration(
+        media_type=media_type,
+        prompt=prompt,
+        file_path=file_path,
+    )
+    db.add(record)
+    db.commit()
+    db.close()

media-gen-api/app/services/image_service.py

@@ -0,0 +1,38 @@
+# app/services/image_service.py
+import os
+from datetime import datetime
+from app.db import SessionLocal
+from app.models import MediaGeneration
+import logging
+logger = logging.getLogger(__name__)
+
+
+def generate_image_file(prompt: str, style: str = "default") -> str:
+    try:
+        # Simulate saving a generated image file
+        filename = f"image_{datetime.now().strftime('%Y%m%d_%H%M%S')}.png"
+        folder = "generated/image"
+        os.makedirs(folder, exist_ok=True)
+    
+        # Placeholder: Simulate image generation by writing prompt text to a file
+        with open(os.path.join(folder, filename), "w") as f:
+            f.write(f"Prompt: {prompt}\nStyle: {style}")
+        logger.info(f"Generated Image: {filename}")
+        return filename
+    except:
+        logger.error(f"Image Geneartion failed: {str(e)}")
+        raise
+
+from app.db import SessionLocal
+from app.models import MediaGeneration
+
+def save_metadata(media_type, prompt, file_path):
+    db = SessionLocal()
+    record = MediaGeneration(
+        media_type=media_type,
+        prompt=prompt,
+        file_path=file_path,
+    )
+    db.add(record)
+    db.commit()
+    db.close()

media-gen-api/app/services/ppt_service.py

@@ -0,0 +1,34 @@
+# app/services/ppt_service.py
+import os
+from datetime import datetime
+from app.db import SessionLocal
+from app.models import MediaGeneration
+import logging
+logger = logging.getLogger(__name__)
+
+def generate_ppt_file(slides: list[dict]) -> str:
+    try:
+        filename = f"ppt_{datetime.now().strftime('%Y%m%d_%H%M%S')}.ppt"
+        folder = "generated/ppt"
+        os.makedirs(folder, exist_ok=True)
+    
+        with open(os.path.join(folder, filename), "w") as f:
+            for i, slide in enumerate(slides, 1):
+                f.write(f"Slide {i}:\nTitle: {slide['title']}\nContent: {slide['content']}\n\n")
+        logger.info(f"Generated PPT: {filename}")
+        return filename
+    except:
+        logger.error(f"PPT Generation failed: {str(e)}")
+        raise
+    
+
+def save_metadata(media_type, prompt, file_path):
+    db = SessionLocal()
+    record = MediaGeneration(
+        media_type=media_type,
+        prompt=prompt,
+        file_path=file_path,
+    )
+    db.add(record)
+    db.commit()
+    db.close()

media-gen-api/app/services/video_service.py

@@ -0,0 +1,35 @@
+# app/services/video_service.py
+import os
+from datetime import datetime
+from app.db import SessionLocal
+from app.models import MediaGeneration
+import logging
+logger = logging.getLogger(__name__)
+
+def generate_video_file(script: str, duration: int = 10) -> str:
+    try:
+        # Simulate saving a generated video file
+        filename = f"video_{datetime.now().strftime('%Y%m%d_%H%M%S')}.mp4"
+        folder = "generated/video"
+        os.makedirs(folder, exist_ok=True)
+    
+        # Placeholder: Simulate video generation by writing script info to a file
+        with open(os.path.join(folder, filename), "w") as f:
+            f.write(f"Script: {script}\nDuration: {duration} seconds")
+        logger.info(f"Generated Video: {filename}")
+        return filename
+    except:
+        logger.error(f"Video generation failed: {str(e)}")
+        raise
+    
+
+def save_metadata(media_type, prompt, file_path):
+    db = SessionLocal()
+    record = MediaGeneration(
+        media_type=media_type,
+        prompt=prompt,
+        file_path=file_path,
+    )
+    db.add(record)
+    db.commit()
+    db.close()

media-gen-api/backend/media_gen.py

@@ -0,0 +1,187 @@
+# ✅ Updated media_gen.py with file logging + UI debug toggle
+import os
+import re
+import logging
+import streamlit as st
+import requests
+from PIL import Image, UnidentifiedImageError
+from io import BytesIO
+from dotenv import load_dotenv
+from moviepy.editor import ImageClip, AudioFileClip
+from elevenlabs import generate, save, set_api_key
+from googletrans import Translator
+from PIL import ImageEnhance, Image
+import tempfile
+
+# Load env vars
+load_dotenv()
+
+# Logging setup
+logging.basicConfig(
+    filename="app.log",
+    level=logging.INFO,
+    format="%(asctime)s [%(levelname)s] %(message)s",
+)
+
+# Constants
+OUTPUT_DIR = "outputs"
+DEFAULT_IMAGE = "assets/fallback.jpg"
+WATERMARK_PATH = "assets/logo_watermark.png"
+UNSPLASH_ACCESS_KEY = os.getenv("UNSPLASH_ACCESS_KEY")
+
+os.makedirs("outputs/audio", exist_ok=True)
+os.makedirs("outputs/images", exist_ok=True)
+os.makedirs("outputs/videos", exist_ok=True)
+
+def translate_text(text, target_lang):
+    return Translator().translate(text, dest=target_lang).text
+
+def sanitize_filename(text):
+    return re.sub(r'\W+', '_', text).lower()[:50]
+
+def apply_watermark(image_path, watermark_path=WATERMARK_PATH):
+    try:
+        base = Image.open(image_path).convert("RGBA")
+        watermark = Image.open(watermark_path).convert("RGBA").resize((100, 100))
+        base.paste(watermark, (base.width - 110, base.height - 110), watermark)
+        base.convert("RGB").save(image_path)
+    except Exception as e:
+        logging.error(f"Watermarking failed: {e}")
+        st.write(f"❌ Watermarking failed: {e}")
+
+def use_fallback_image(prompt, add_watermark=False):
+    try:
+        fallback_path = DEFAULT_IMAGE
+        output_path = f"outputs/images/{sanitize_filename(prompt)}.jpg"
+        with Image.open(fallback_path) as img:
+            img.save(output_path)
+        if add_watermark:
+            apply_watermark(output_path)
+        return output_path
+    except UnidentifiedImageError:
+        logging.error("Could not open fallback image.")
+        st.write("❌ Could not open fallback image.")
+        return None
+
+def generate_gtts_fallback(prompt, output_path, lang="en", debug_mode=False):
+    try:
+        from gtts import gTTS
+        tts = gTTS(text=prompt, lang=lang)
+        tts.save(output_path)
+        logging.info(f"gTTS fallback audio saved to {output_path}")
+        if debug_mode:
+            st.write(f"✅ Fallback audio (gTTS) saved to {output_path}")
+        return output_path
+    except Exception as e:
+        logging.error(f"gTTS fallback failed: {e}")
+        st.write(f"❌ gTTS fallback failed: {str(e)}")
+        return None
+
+
+def generate_image(prompt, file_tag, add_watermark=False, dark_mode=False, debug_mode=False):
+    try:
+        # Enhance prompt if dark mode is enabled
+        if dark_mode:
+            prompt += " at night, dark theme, low light, moody lighting"
+
+        url = f"https://api.unsplash.com/photos/random?query={requests.utils.quote(prompt)}&client_id={UNSPLASH_ACCESS_KEY}"
+        response = requests.get(url, timeout=10)
+        response.raise_for_status()
+        image_url = response.json()["urls"]["regular"]
+        image_response = requests.get(image_url, timeout=10)
+        image_response.raise_for_status()
+
+        output_path = f"outputs/images/{sanitize_filename(prompt)}.jpg"
+        img = Image.open(BytesIO(image_response.content))
+        img.convert("RGB").save(output_path)
+
+        if add_watermark:
+            apply_watermark(output_path)
+
+        return output_path
+
+    except Exception as e:
+        logging.error(f"Image generation failed: {e}")
+        st.write("🔁 Unsplash failed. Using fallback.")
+        st.write(f"❌ Image generation failed: {e}")
+        return use_fallback_image(prompt, add_watermark=add_watermark)
+
+
+# ✅ Updated generate_audio with proper language handling
+
+def generate_audio(prompt, output_path, debug_mode=False, lang="en"):
+    try:
+        api_key = os.getenv("ELEVEN_API_KEY") or st.secrets.get("ELEVEN_API_KEY", None)
+
+        # Use gTTS for non-English languages
+        if lang != "en":
+            if debug_mode:
+                st.write(f"🌐 Non-English language selected: {lang}. Using gTTS.")
+            return generate_gtts_fallback(prompt, output_path, lang=lang, debug_mode=debug_mode)
+
+        if api_key:
+            if debug_mode:
+                st.write(f"✅ ELEVEN_API_KEY loaded: {api_key[:4]}...****")
+
+            set_api_key(api_key)
+            if debug_mode:
+                st.write(f"🎧 Generating audio for prompt: {prompt}")
+
+            try:
+                audio = generate(text=prompt, voice="Aria", model="eleven_monolingual_v1")
+                save(audio, output_path)
+                logging.info(f"Audio saved successfully to {output_path}")
+
+                if debug_mode:
+                    st.write(f"🔍 File exists after save? {os.path.exists(output_path)}")
+                    st.write(f"✅ Audio saved successfully to {output_path}")
+                return output_path
+
+            except Exception as e:
+                logging.warning(f"ElevenLabs failed: {e}")
+                if debug_mode:
+                    st.write(f"⚠️ ElevenLabs failed: {str(e)}")
+                    st.write("🔁 Falling back to gTTS...")
+                return generate_gtts_fallback(prompt, output_path, lang=lang, debug_mode=debug_mode)
+
+        else:
+            logging.warning("ELEVEN_API_KEY not found")
+            if debug_mode:
+                st.write("❌ ELEVEN_API_KEY not found. Falling back to gTTS.")
+            return generate_gtts_fallback(prompt, output_path, lang=lang, debug_mode=debug_mode)
+
+    except Exception as e:
+        logging.error(f"Exception during audio generation setup: {e}")
+        if debug_mode:
+            st.write(f"❌ Exception during audio generation setup: {str(e)}")
+            st.write("🔁 Falling back to gTTS...")
+        return generate_gtts_fallback(prompt, output_path, lang=lang, debug_mode=debug_mode)
+  
+
+def generate_video(prompt, image_path, audio_path, output_path, add_watermark=False, dark_mode=False):
+    try:
+        # If dark_mode, darken the image temporarily
+        if dark_mode:
+            with Image.open(image_path) as img:
+                enhancer = ImageEnhance.Brightness(img)
+                darker_img = enhancer.enhance(0.5)  # Reduce brightness to 50%
+                
+                # Save to a temporary file
+                with tempfile.NamedTemporaryFile(delete=False, suffix=".jpg") as tmp:
+                    temp_image_path = tmp.name
+                    darker_img.save(temp_image_path)
+                    image_path = temp_image_path
+
+        audio_clip = AudioFileClip(audio_path)
+        image_clip = ImageClip(image_path).set_duration(audio_clip.duration).resize(height=720)
+        video = image_clip.set_audio(audio_clip)
+
+        output_path = f"outputs/videos/{sanitize_filename(prompt)}.mp4"
+        video.write_videofile(output_path, fps=24, codec="libx264", audio_codec="aac", verbose=False, logger=None)
+        return output_path
+
+    except Exception as e:
+        logging.error(f"Video generation failed: {e}")
+        st.write(f"❌ Video generation failed: {e}")
+        return None
+

media-gen-api/docs/index.html

@@ -0,0 +1,62 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+  <meta charset="UTF-8" />
+  <meta name="viewport" content="width=device-width, initial-scale=1.0"/>
+  <meta name="description" content="OSN Media Generator - Built by O.S.Narayana. Create media content using AI with no ads or in-app purchases."/>
+  <title>OSN Media Generator</title>
+  <style>
+    body {
+      font-family: Arial, sans-serif;
+      background-color: #f3f4f6;
+      color: #111827;
+      text-align: center;
+      padding: 50px 20px;
+    }
+    .container {
+      max-width: 600px;
+      margin: auto;
+      background: #ffffff;
+      padding: 30px;
+      border-radius: 12px;
+      box-shadow: 0 4px 12px rgba(0,0,0,0.1);
+    }
+    h1 {
+      color: #1f2937;
+    }
+    a {
+      color: #2563eb;
+      text-decoration: none;
+    }
+    a:hover {
+      text-decoration: underline;
+    }
+    .footer {
+      margin-top: 40px;
+      font-size: 0.9em;
+      color: #6b7280;
+    }
+  </style>
+</head>
+<body>
+  <div class="container">
+    <h1>🎮 OSN Media Generator</h1>
+    <p>
+      Welcome to the official homepage of <strong>OSN Media Generator</strong>, created by <strong>O.S.Narayana</strong>.
+    </p>
+    <p>
+      This app helps you generate audio, images, and videos using AI tools like ElevenLabs, Unsplash, and Streamlit.
+    </p>
+    <p>
+      👉 Try the app now: <br>
+      <a href="https://osnarayana-media-generator.streamlit.app/" target="_blank">Launch OSN Media Generator</a>
+    </p>
+    <p>
+      📄 View our <a href="privacy.html" target="_blank">Privacy Policy</a>
+    </p>
+    <div class="footer">
+      &copy; 2025 OSN Media. Built with 💗 by O.S.Narayana.
+    </div>
+  </div>
+</body>
+</html>

media-gen-api/docs/privacy.html

@@ -0,0 +1,68 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+    <meta charset="UTF-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1.0">
+    <title>Privacy Policy - OSN Media Generator</title>
+    <meta http-equiv="Cache-Control" content="no-cache, no-store, must-revalidate" />
+    <meta http-equiv="Pragma" content="no-cache" />
+    <meta http-equiv="Expires" content="0" />
+    <style>
+        body {
+            font-family: Arial, sans-serif;
+            margin: 2rem;
+            padding: 0;
+            background-color: #f9f9f9;
+            color: #333;
+        }
+        h1, h2, h3 { color: #222; }
+        a { color: #1a73e8; text-decoration: none; }
+        a:hover { text-decoration: underline; }
+        code { background-color: #eee; padding: 2px 4px; border-radius: 4px; }
+       
+    </style>
+</head>
+<body>
+    <h1>🔐 Privacy Policy for OSN Media Generator</h1>
+
+<p>Effective Date: July 19, 2025</p>
+
+<p>Your privacy is important to us. This app is built to prioritize user data security and transparency.</p>
+
+<h2>1. What We Collect</h2>
+
+<ul>
+<li><strong>No personal data</strong> is collected, stored, or shared by this application.</li>
+<li>All prompts, audio, and media are <strong>processed locally</strong> or via <strong>user-provided API keys</strong> to external services (e.g., ElevenLabs, Unsplash).</li>
+</ul>
+
+<h2>2. API Usage</h2>
+
+<ul>
+<li>If you use external services (like ElevenLabs or Unsplash), you are subject to their respective <a href="https://www.elevenlabs.io/terms">Terms of Service</a> and <a href="https://www.elevenlabs.io/privacy">Privacy Policies</a>.</li>
+<li>Your API keys are stored <strong>locally</strong> in your environment file (<code>.env</code>) and never uploaded.</li>
+</ul>
+
+<h2>3. Data Storage</h2>
+
+<ul>
+<li>Generated images, audio, and videos are stored <strong>only on your local machine</strong> under the <code>outputs/</code> directory.</li>
+<li>You can delete any generated content at your discretion.</li>
+</ul>
+
+<h2>4. Analytics &amp; Ads</h2>
+
+<ul>
+<li>This app contains <strong>no ads</strong>, <strong>no tracking</strong>, and <strong>no analytics</strong>.</li>
+</ul>
+
+<h2>5. Changes to Policy</h2>
+
+<p>Any updates to this policy will be reflected in this file on the <a href="https://github.com/sivaogeti/osnarayana-media-generator">GitHub repository</a>.</p>
+
+<h2>6. Contact</h2>
+
+<p>For questions or issues, contact: <strong>[email protected]</strong></p>
+
+</body>
+</html>

media-gen-api/nixpacks.toml

@@ -0,0 +1,19 @@
+[phases.setup]
+nixpkgs = [
+  "python311",
+  "zlib",
+  "libjpeg",
+  "gcc",
+  "pkg-config"
+]
+
+[phases.build]
+cmds = [
+  "python -m venv /opt/venv",
+  ". /opt/venv/bin/activate",
+  "pip install --upgrade pip",
+  "pip install -r requirements.txt"
+]
+
+[start]
+cmd = "streamlit run app.py --server.port 8000 --server.address 0.0.0.0"

media-gen-api/openapi.json

@@ -0,0 +1 @@
+{"openapi":"3.1.0","info":{"title":"Media Generator API","description":"Generate audio, video, image, and PPT from text","version":"1.0.0"},"paths":{"/api/v1/audio/generate":{"post":{"tags":["Audio"],"summary":"Generate Audio Endpoint","operationId":"generate_audio_endpoint_api_v1_audio_generate_post","requestBody":{"content":{"application/json":{"schema":{"$ref":"#/components/schemas/AudioInput"}}},"required":true},"responses":{"200":{"description":"Successful Response","content":{"application/json":{"schema":{}}}},"422":{"description":"Validation Error","content":{"application/json":{"schema":{"$ref":"#/components/schemas/HTTPValidationError"}}}}},"security":[{"HTTPBearer":[]}]}},"/api/v1/video/generate":{"post":{"tags":["Video"],"summary":"Generate Video","operationId":"generate_video_api_v1_video_generate_post","requestBody":{"content":{"application/json":{"schema":{"$ref":"#/components/schemas/VideoInput"}}},"required":true},"responses":{"200":{"description":"Successful Response","content":{"application/json":{"schema":{}}}},"422":{"description":"Validation Error","content":{"application/json":{"schema":{"$ref":"#/components/schemas/HTTPValidationError"}}}}},"security":[{"HTTPBearer":[]}]}},"/api/v1/image/generate":{"post":{"tags":["Image"],"summary":"Generate Image","operationId":"generate_image_api_v1_image_generate_post","requestBody":{"content":{"application/json":{"schema":{"$ref":"#/components/schemas/ImageInput"}}},"required":true},"responses":{"200":{"description":"Successful Response","content":{"application/json":{"schema":{}}}},"422":{"description":"Validation Error","content":{"application/json":{"schema":{"$ref":"#/components/schemas/HTTPValidationError"}}}}},"security":[{"HTTPBearer":[]}]}},"/api/v1/ppt/generate":{"post":{"tags":["PPT"],"summary":"Generate Ppt","operationId":"generate_ppt_api_v1_ppt_generate_post","requestBody":{"content":{"application/json":{"schema":{"$ref":"#/components/schemas/PPTInput"}}},"required":true},"responses":{"200":{"description":"Successful Response","content":{"application/json":{"schema":{}}}},"422":{"description":"Validation Error","content":{"application/json":{"schema":{"$ref":"#/components/schemas/HTTPValidationError"}}}}},"security":[{"HTTPBearer":[]}]}},"/api/v1/metrics/evaluate/bleu":{"post":{"tags":["Metrics"],"summary":"Compute Bleu","operationId":"compute_bleu_api_v1_metrics_evaluate_bleu_post","parameters":[{"name":"reference","in":"query","required":true,"schema":{"type":"string","title":"Reference"}},{"name":"candidate","in":"query","required":true,"schema":{"type":"string","title":"Candidate"}}],"responses":{"200":{"description":"Successful Response","content":{"application/json":{"schema":{}}}},"422":{"description":"Validation Error","content":{"application/json":{"schema":{"$ref":"#/components/schemas/HTTPValidationError"}}}}},"security":[{"HTTPBearer":[]}]}},"/api/v1/metrics/evaluate/clipscore":{"post":{"tags":["Metrics"],"summary":"Compute Clip Score","operationId":"compute_clip_score_api_v1_metrics_evaluate_clipscore_post","parameters":[{"name":"reference","in":"query","required":true,"schema":{"type":"string","title":"Reference"}},{"name":"candidate","in":"query","required":true,"schema":{"type":"string","title":"Candidate"}}],"responses":{"200":{"description":"Successful Response","content":{"application/json":{"schema":{}}}},"422":{"description":"Validation Error","content":{"application/json":{"schema":{"$ref":"#/components/schemas/HTTPValidationError"}}}}},"security":[{"HTTPBearer":[]}]}},"/api/v1/download/":{"get":{"tags":["Download"],"summary":"Download File","operationId":"download_file_api_v1_download__get","parameters":[{"name":"file_path","in":"query","required":true,"schema":{"type":"string","description":"Relative path from project root","title":"File Path"},"description":"Relative path from project root"}],"responses":{"200":{"description":"Successful Response","content":{"application/json":{"schema":{}}}},"422":{"description":"Validation Error","content":{"application/json":{"schema":{"$ref":"#/components/schemas/HTTPValidationError"}}}}},"security":[{"HTTPBearer":[]}]}}},"components":{"schemas":{"AudioInput":{"properties":{"text":{"type":"string","title":"Text"},"voice":{"type":"string","title":"Voice","default":"default"},"language":{"type":"string","title":"Language","default":"en"}},"type":"object","required":["text"],"title":"AudioInput"},"HTTPValidationError":{"properties":{"detail":{"items":{"$ref":"#/components/schemas/ValidationError"},"type":"array","title":"Detail"}},"type":"object","title":"HTTPValidationError"},"ImageInput":{"properties":{"prompt":{"type":"string","title":"Prompt"},"style":{"type":"string","title":"Style","default":"default"}},"type":"object","required":["prompt"],"title":"ImageInput"},"PPTInput":{"properties":{"slides":{"items":{"$ref":"#/components/schemas/Slide"},"type":"array","title":"Slides"}},"type":"object","required":["slides"],"title":"PPTInput"},"Slide":{"properties":{"title":{"type":"string","title":"Title"},"content":{"type":"string","title":"Content"}},"type":"object","required":["title","content"],"title":"Slide"},"ValidationError":{"properties":{"loc":{"items":{"anyOf":[{"type":"string"},{"type":"integer"}]},"type":"array","title":"Location"},"msg":{"type":"string","title":"Message"},"type":{"type":"string","title":"Error Type"}},"type":"object","required":["loc","msg","type"],"title":"ValidationError"},"VideoInput":{"properties":{"script":{"type":"string","title":"Script"},"duration":{"type":"integer","title":"Duration","default":10}},"type":"object","required":["script"],"title":"VideoInput"}},"securitySchemes":{"HTTPBearer":{"type":"http","scheme":"bearer"}}}}

media-gen-api/packages.txt

@@ -0,0 +1,3 @@
+ffmpeg
+libjpeg-dev
+zlib1g-dev

media-gen-api/requirements.txt

@@ -0,0 +1,19 @@
+streamlit
+requests
+python-dotenv
+gtts
+moviepy==1.0.3
+google-api-python-client
+google-auth-httplib2
+google-auth-oauthlib
+google-auth
+pillow==9.5.0
+googletrans==4.0.0-rc1 
+elevenlabs
+deep_translator
+gTTS
+sentry-sdk
+scikit-learn
+nltk
+sentence_transformers
+sqlalchemy

media-gen-api/scripts/preload_content.py

@@ -0,0 +1,17 @@
+import os
+import shutil
+from datetime import datetime
+
+SOURCE_DIR = "generated"
+ARCHIVE_DIR = f"archive_{datetime.now().strftime('%Y%m%d_%H%M')}"
+
+def archive_all():
+    if not os.path.exists(SOURCE_DIR):
+        print("No content to archive.")
+        return
+
+    shutil.copytree(SOURCE_DIR, ARCHIVE_DIR)
+    print(f"Archived to {ARCHIVE_DIR}")
+
+if __name__ == "__main__":
+    archive_all()

media-gen-api/streamlit_ui.py

@@ -0,0 +1,75 @@
+# streamlit_ui.py
+import streamlit as st
+import requests
+import base64
+
+st.set_page_config(page_title="AI Media Generator", layout="wide")
+st.title("🎙️🖼️🎞️ AI Media Generator")
+
+API_BASE = "http://localhost:8000"  # change if deployed
+
+# Helper to play audio/video in Streamlit
+def render_media(file_bytes, media_type, label):
+    b64 = base64.b64encode(file_bytes).decode()
+    if media_type == "audio":
+        st.audio(f"data:audio/wav;base64,{b64}", format="audio/wav")
+    elif media_type == "video":
+        st.video(f"data:video/mp4;base64,{b64}")
+    elif media_type == "image":
+        st.image(file_bytes, caption=label, use_column_width=True)
+
+# Sidebar inputs
+st.sidebar.header("🛠️ Settings")
+TOKEN = st.sidebar.text_input("🔑 API Token", type="password")
+HEADERS = {"Authorization": f"Bearer {TOKEN}"} if TOKEN else {}
+
+# Tabs
+tab = st.sidebar.radio("Select Task", ["Text to Audio", "Text to Image", "Text to Video"])
+
+if tab == "Text to Audio":
+    st.subheader("🎤 Text to Audio")
+    text = st.text_area("Enter text")
+    voice = st.selectbox("Choose voice/language", ["en-US", "hi-IN", "te-IN", "ta-IN"])
+    
+    if st.button("🔊 Generate Audio"):
+        with st.spinner("Generating audio..."):
+            r = requests.post(f"{API_BASE}/audio/generate", json={"text": text, "voice": voice}, headers=HEADERS)
+            if r.status_code == 200:
+                render_media(r.content, "audio", "Generated Audio")
+            else:
+                st.error(f"❌ Failed: {r.json().get('detail')}")
+
+elif tab == "Text to Image":
+    st.subheader("🖼️ Text to Image")
+    prompt = st.text_area("Enter image prompt")
+    model = st.selectbox("Choose model", ["sdxl", "deepfloyd", "kandinsky"])
+
+    if st.button("🧠 Generate Image"):
+        with st.spinner("Generating image..."):
+            r = requests.post(f"{API_BASE}/image/generate", json={"prompt": prompt, "model": model}, headers=HEADERS)
+            if r.status_code == 200:
+                render_media(r.content, "image", "Generated Image")
+            else:
+                st.error(f"❌ Failed: {r.json().get('detail')}")
+
+elif tab == "Text to Video":
+    st.subheader("🎞️ Text to Video")
+    prompt = st.text_area("Enter video prompt")
+    tone = st.selectbox("Tone", ["formal", "casual", "emotional", "documentary"])
+    domain = st.selectbox("Domain", ["health", "education", "governance", "entertainment"])
+    environment = st.selectbox("Environment", ["urban", "rural", "nature", "futuristic"])
+
+    if st.button("🎬 Generate Video"):
+        with st.spinner("Generating video..."):
+            r = requests.post(
+                f"{API_BASE}/video/generate",
+                json={"prompt": prompt, "tone": tone, "domain": domain, "environment": environment},
+                headers=HEADERS
+            )
+            if r.status_code == 200:
+                render_media(r.content, "video", "Generated Video")
+            else:
+                st.error(f"❌ Failed: {r.json().get('detail')}")
+
+st.sidebar.markdown("---")
+st.sidebar.info("Built with ❤️ for AI GovTech Challenge 2025")