Add Gemini API key and model configuration to UI

README.md

@@ -11,4 +11,72 @@ tags:
 - tool
 ---
 
+# YouTube Transcript Summarizer and Blog Content Generator
+
+This tool extracts transcripts from YouTube videos, summarizes them using Google's Gemini AI, generates relevant images using Hugging Face models, and creates a formatted DOCX document with the content.
+
+## Features
+
+- Extract transcripts from YouTube videos with automatic language detection
+- Summarize transcripts using Google's Gemini AI models
+- Generate relevant images based on the summary content
+- Create or update DOCX documents with the transcript, summary, and images
+- Support for multiple languages (summarizes in the same language as the transcript)
+- Configurable AI models and API keys directly from the UI
+
+## Setup and Usage
+
+1. **API Keys**:
+   - **Hugging Face API Key**: Required for image generation
+   - **Gemini API Key**: Required for transcript summarization
+   - Both keys can be entered directly in the UI
+
+2. **Model Configuration**:
+   - Select from different Gemini models:
+     - gemini-2.0-flash (default)
+     - gemini-1.5-pro
+     - gemini-1.5-flash
+
+3. **Input**:
+   - Enter a YouTube video URL
+   - Optionally provide an existing DOCX file to update
+
+4. **Output**:
+   - Extracted transcript
+   - Generated summary
+   - Generated image
+   - DOCX document with all content
+
+## Environment Variables
+
+You can optionally set API keys in a `.env` file:
+
+```
+HF_API_KEY=your_hugging_face_api_key
+GEMINI_API_KEY=your_gemini_api_key
+```
+
+## Requirements
+
+- Python 3.8+
+- Required packages are listed in `requirements.txt`
+
+## Installation
+
+```bash
+# Clone the repository
+git clone https://github.com/yourusername/tool-YoutubeTranscript-blog.git
+cd tool-YoutubeTranscript-blog
+
+# Create a virtual environment
+python -m venv .venv
+source .venv/bin/activate  # On Windows: .venv\Scripts\activate
+
+# Install dependencies
+pip install -r requirements.txt
+
+# Run the application
+python app.py
+```
+
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -10,9 +10,13 @@ youtube_tool = YouTubeTranscriptExtractor()
 docx_tool = TranscriptToDocx()
 #summarizer_tool = TranscriptSummarizer()
 
-def process_youtube_video(video_url, hf_api_key, existing_docx_path=None):
+def process_youtube_video(video_url, hf_api_key, gemini_api_key, model_name, existing_docx_path=None):
     # Initialize tools
-    summarizer_tool = TranscriptSummarizer(hf_api_key=hf_api_key)
+    summarizer_tool = TranscriptSummarizer(
+        hf_api_key=hf_api_key,
+        gemini_api_key=gemini_api_key,
+        model_name=model_name
+    )
 
     # Get video title
     from pytubefix import YouTube
@@ -86,15 +90,30 @@ with gr.Blocks() as demo:
     gr.Markdown("# YouTube Transcript Summarizer and Blog Content Generator")
     gr.Markdown("Enter a YouTube video URL and Hugging Face API Key to extract the transcript, summarize it, and generate blog content with an image and DOCX file. Optionally, you can provide an existing DOCX file to update.")
 
-    # Check if Gemini API key is set
-    gemini_api_key = os.getenv("GEMINI_API_KEY")
-    if not gemini_api_key or gemini_api_key == "your_gemini_api_key_here":
-        gr.Markdown("⚠️ **Warning**: Gemini API key is not set in the .env file. Please add your Gemini API key to the .env file to use the summarization feature.")
+    # Information about API keys
+    gr.Markdown("### API Keys and Model Configuration")
+    gr.Markdown("You need to provide both a Hugging Face API key (for image generation) and a Gemini API key (for summarization). You can now enter your Gemini API key directly in the UI or set it in the .env file.")
 
     with gr.Row():
         with gr.Column():
             video_url = gr.Textbox(label="YouTube Video URL")
+
+            # API Keys
+            gr.Markdown("#### API Keys")
             hf_api_key = gr.Textbox(label="Hugging Face API Key", type="password")
+            gemini_api_key = gr.Textbox(
+                label="Gemini API Key",
+                type="password",
+                value=os.getenv("GEMINI_API_KEY", "")
+            )
+
+            # Model Selection
+            model_name = gr.Dropdown(
+                label="Gemini Model",
+                choices=["gemini-2.0-flash", "gemini-1.5-pro", "gemini-1.5-flash"],
+                value="gemini-2.0-flash"
+            )
+
             existing_docx = gr.File(label="Existing DOCX file (optional)", file_types=[".docx"])
             submit_btn = gr.Button("Process Video")
 
@@ -106,7 +125,7 @@ with gr.Blocks() as demo:
 
     submit_btn.click(
         fn=process_youtube_video,
-        inputs=[video_url, hf_api_key, existing_docx],
+        inputs=[video_url, hf_api_key, gemini_api_key, model_name, existing_docx],
         outputs=[transcript_output, summary_output, image_output, docx_output]
     )
 

tool.py

@@ -24,15 +24,17 @@ class TranscriptSummarizer(Tool):
     }
     output_type = "string"
 
-    def __init__(self, *args, hf_api_key: str = None, **kwargs):
+    def __init__(self, *args, hf_api_key: str = None, gemini_api_key: str = None, model_name: str = 'gemini-2.0-flash', **kwargs):
         super().__init__(*args, **kwargs)
-        # Get Gemini API key from environment variables
-        gemini_api_key = os.getenv("GEMINI_API_KEY")
-        if gemini_api_key:
+        # Use provided Gemini API key or get from environment variables
+        self.gemini_api_key = gemini_api_key or os.getenv("GEMINI_API_KEY")
+        self.model_name = model_name
+
+        if self.gemini_api_key:
             # Configure the Gemini API
-            genai.configure(api_key=gemini_api_key)
-            # Set up the model
-            self.gemini_model = genai.GenerativeModel('gemini-2.0-flash')
+            genai.configure(api_key=self.gemini_api_key)
+            # Set up the model with the specified model name
+            self.gemini_model = genai.GenerativeModel(self.model_name)
         else:
             self.gemini_model = None
 
@@ -106,7 +108,7 @@ class TranscriptSummarizer(Tool):
                 return "Hugging Face API key is required for image generation. Please provide it in the input field."
 
             if not self.gemini_model:
-                return "Gemini API key is required for summarization. Please add it to your .env file."
+                return "Gemini API key is required for summarization. Please provide it in the input field."
 
             transcript_length = len(transcript)