Add YouTube transcript summarization tool with Gradio interface and image generation which will further use for blog writing.

.gitignore

@@ -0,0 +1,18 @@
+.ipynb_checkpoints/
+__pycache__/
+*.py[cod]
+*$py.class
+.env
+venv/
+__pycache__/
+*.pyc
+*.pyo
+*$py.class
+*~
+*.swp
+*.log
+/instance
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class

app.py

@@ -1,7 +1,29 @@
-from smolagents import launch_gradio_demo
-from typing import Optional
-from tool import YouTubeTranscriptExtractor
+import gradio as gr
+from tool import YouTubeTranscriptExtractor, TranscriptSummarizer
 
-tool = YouTubeTranscriptExtractor()
+youtube_tool = YouTubeTranscriptExtractor()
+summarizer_tool = TranscriptSummarizer()
 
-launch_gradio_demo(tool)
+def process_youtube_video(video_url):
+    transcript = youtube_tool.forward(video_url=video_url)
+    summary_and_blog = summarizer_tool.forward(transcript=transcript)
+    try:
+        summary, image_url = summary_and_blog.split("\n\nImage URL: ")
+    except ValueError:
+        summary = summary_and_blog
+        image_url = None
+    return transcript, summary, image_url
+
+iface = gr.Interface(
+    fn=process_youtube_video,
+    inputs="text",
+    outputs=[
+        gr.Textbox(label="Transcript"),
+        gr.Textbox(label="Summary and Blog Content"),
+        gr.Image(label="Generated Image", image_mode="RGBA")
+    ],
+    title="YouTube Transcript Summarizer and Blog Content Generator",
+    description="Enter a YouTube video URL to extract the transcript, summarize it, and generate blog content with an image."
+)
+
+iface.launch()

requirements.txt

@@ -1,2 +1,9 @@
 smolagents
 pytubefix
+gradio
+transformers
+torch
+torchvision
+python-dotenv
+requests
+Pillow

tool.py

@@ -1,5 +1,42 @@
 from smolagents.tools import Tool
-import pytubefix
+from typing import Optional
+import os
+from transformers import pipeline
+import requests
+import io
+from PIL import Image
+from dotenv import load_dotenv
+
+load_dotenv()
+
+class TranscriptSummarizer(Tool):
+    description = "Summarizes a transcript and generates blog content using the transformers library and Hugging Face API for image generation."
+    name = "transcript_summarizer"
+    inputs = {'transcript': {'type': 'string', 'description': 'The transcript to summarize.'}}
+    output_type = "string"
+
+    def __init__(self, *args, **kwargs):
+        super().__init__(*args, **kwargs)
+        self.summarizer = pipeline("summarization", model="facebook/bart-large-cnn")
+        self.api_url = "https://api-inference.huggingface.co/models/ZB-Tech/Text-to-Image"
+        self.headers = {"Authorization": f"Bearer {os.getenv('HF_API_KEY')}"}
+
+    def query(self, payload):
+        response = requests.post(self.api_url, headers=self.headers, json=payload)
+        return response.content
+
+    def forward(self, transcript: str) -> str:
+        try:
+            summary = self.summarizer(transcript, max_length=2000, min_length=750, do_sample=False)[0]['summary_text']
+            key_entities = summary.split()[:100]  # Extract the first 100 words
+            image_prompt = f"Generate an image related to: {' '.join(key_entities)}, professional style"
+            image_bytes = self.query({"inputs": image_prompt})
+            image = Image.open(io.BytesIO(image_bytes))
+            image_url = "Images/image.jpg"  # Placeholder, as we can't directly pass PIL Image to Gradio
+            image.save(image_url) # Save the image to a file
+            return f"{summary}\n\nImage URL: {image_url}" # Return the file path
+        except Exception as e:
+            return f"An unexpected error occurred: {str(e)}"
 
 class YouTubeTranscriptExtractor(Tool):
     description = "Extracts the transcript from a YouTube video."
@@ -8,30 +45,28 @@ class YouTubeTranscriptExtractor(Tool):
     output_type = "string"
 
     def forward(self, video_url: str) -> str:
-
         try:
-          from pytubefix import YouTube
-          # Create a YouTube object
-          yt = YouTube(video_url)
-          lang='en'
-
-          # Get the video transcript
-          if lang in yt.captions:
-              transcript = yt.captions['en'].generate_srt_captions()
-
-          else:
-              transcript = yt.captions.all()[0].generate_srt_captions()
-              lang=yt.captions.all()[0].code
-
-          return lang + "transcript : " + transcript
-        # return transcript
+            from pytubefix import YouTube
+            # Create a YouTube object
+            yt = YouTube(video_url)
+            lang='en'
+            # Get the video transcript
+            if lang in yt.captions:
+                transcript = yt.captions['en'].generate_srt_captions()
+            else:
+                transcript = yt.captions.all()[0].generate_srt_captions()
+                lang = yt.captions.all()[0].code
 
+            # Clean up the transcript by removing timestamps and line numbers
+            cleaned_transcript = ""
+            for line in transcript.splitlines():
+                if not line.strip().isdigit() and "-->" not in line:
+                    cleaned_transcript += line + "\n"
 
+            print("transcript : ", cleaned_transcript)
+            return cleaned_transcript
         except Exception as e:
-
-          return f"An unexpected error occurred: {str(e)}"
-
+            return f"An unexpected error occurred: {str(e)}"
 
     def __init__(self, *args, **kwargs):
         self.is_initialized = False
-