Create app.py

app.py

@@ -0,0 +1,128 @@
+from fastapi import FastAPI, HTTPException
+from fastapi.responses import FileResponse
+from gtts import gTTS
+import cv2
+import numpy as np
+from PIL import Image, ImageDraw, ImageFont
+import os
+
+app = FastAPI()
+
+# Function to split the script into smaller chunks
+def split_script(script: str, max_words: int = 50):
+    words = script.split()
+    chunks = [" ".join(words[i:i + max_words]) for i in range(0, len(words), max_words)]
+    return chunks
+
+# Function to create a video segment from a script chunk
+def create_video_segment(script_chunk: str, background_color: str, text_color: str, font_size: int):
+    try:
+        # Step 1: Convert script chunk to audio using gTTS
+        tts = gTTS(script_chunk)
+        audio_file = "output_audio.mp3"
+        tts.save(audio_file)
+
+        # Step 2: Create a blank image with text
+        width, height = 1280, 720  # HD resolution
+        background_color_rgb = tuple(int(background_color.lstrip("#")[i:i+2], 16) for i in (0, 2, 4))
+        text_color_rgb = tuple(int(text_color.lstrip("#")[i:i+2], 16) for i in (0, 2, 4))
+
+        # Create a blank image with the background color
+        image = Image.new("RGB", (width, height), background_color_rgb)
+        draw = ImageDraw.Draw(image)
+
+        # Load a font (you can use a default font or provide a .ttf file)
+        try:
+            font = ImageFont.truetype("arial.ttf", font_size)
+        except IOError:
+            font = ImageFont.load_default()
+
+        # Add text to the image
+        text_width, text_height = draw.textsize(script_chunk, font=font)
+        text_x = (width - text_width) // 2
+        text_y = (height - text_height) // 2
+        draw.text((text_x, text_y), script_chunk, font=font, fill=text_color_rgb)
+
+        # Convert the image to a numpy array for OpenCV
+        frame = np.array(image)
+        frame = cv2.cvtColor(frame, cv2.COLOR_RGB2BGR)
+
+        # Step 3: Create a video segment with the image and audio
+        video_segment_file = f"video_segment_{len(os.listdir())}.mp4"
+        fps = 24
+        fourcc = cv2.VideoWriter_fourcc(*"mp4v")
+        video_writer = cv2.VideoWriter(video_segment_file, fourcc, fps, (width, height))
+
+        # Calculate the number of frames based on audio duration
+        audio_duration = len(script_chunk.split()) * 0.5  # Approximate duration (adjust as needed)
+        num_frames = int(audio_duration * fps)
+
+        # Write the frames to the video
+        for _ in range(num_frames):
+            video_writer.write(frame)
+
+        # Release the video writer
+        video_writer.release()
+
+        # Step 4: Add audio to the video segment using ffmpeg (if available)
+        if os.path.exists(audio_file):
+            final_video_segment_file = f"final_{video_segment_file}"
+            os.system(f"ffmpeg -i {video_segment_file} -i {audio_file} -c:v copy -c:a aac {final_video_segment_file}")
+            os.remove(video_segment_file)
+            os.remove(audio_file)
+            return final_video_segment_file
+        else:
+            return video_segment_file
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))
+
+# Function to combine video segments into a single video
+def combine_video_segments(video_segment_files: list, output_file: str = "final_output_video.mp4"):
+    try:
+        with open("video_list.txt", "w") as f:
+            for segment in video_segment_files:
+                f.write(f"file '{segment}'\n")
+
+        # Combine video segments using ffmpeg
+        os.system(f"ffmpeg -f concat -safe 0 -i video_list.txt -c copy {output_file}")
+
+        # Clean up video segments
+        for segment in video_segment_files:
+            os.remove(segment)
+        os.remove("video_list.txt")
+
+        return output_file
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))
+
+# API Endpoint to generate video
+@app.post("/generate-video")
+async def generate_video(script: str, background_color: str = "#000000", text_color: str = "#FFFFFF", font_size: int = 50):
+    if not script:
+        raise HTTPException(status_code=400, detail="Script cannot be empty")
+
+    try:
+        # Step 1: Split the script into smaller chunks
+        script_chunks = split_script(script)
+
+        # Step 2: Generate video segments for each chunk
+        video_segment_files = []
+        for chunk in script_chunks:
+            video_segment = create_video_segment(chunk, background_color, text_color, font_size)
+            video_segment_files.append(video_segment)
+
+        # Step 3: Combine video segments into a single video
+        final_video_file = combine_video_segments(video_segment_files)
+
+        # Step 4: Return the final video file
+        if not os.path.exists(final_video_file):
+            raise HTTPException(status_code=500, detail="Failed to generate video")
+
+        return FileResponse(final_video_file, media_type="video/mp4", filename="generated_video.mp4")
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))
+
+# Run the application
+if __name__ == "__main__":
+    import uvicorn
+    uvicorn.run(app, host="0.0.0.0", port=8000)