Added WhisperX support, demo feature, and updated Docker image tags

Author: laurentftech

Dockerfile

@@ -1,23 +1,51 @@
-# Use an official Python runtime as a parent image
-FROM python:3.11-slim-bookworm
+# --- Stage 1: Base dependencies ---
+FROM python:3.11-slim-bookworm AS base
 
 # Set the working directory in the container
 WORKDIR /app
 
 # Install FFmpeg (required for audio processing)
+# Also install git for whisperx dependencies
 RUN apt-get update && apt-get install -y --no-install-recommends \
     ffmpeg \
+    git \
     && rm -rf /var/lib/apt/lists/*
 
-# Copy the requirements file into the container at /app
+# Copy the core requirements file
 COPY requirements.txt .
 
-# Install any needed packages specified in requirements.txt
+# Install core dependencies
 RUN pip install --no-cache-dir -r requirements.txt
 
-# Copy the rest of the application code into the container
+# Copy the rest of the application code
 COPY . .
 
+# --- Stage 2: Build with WhisperX support ---
+FROM base AS with_whisperx
+
+# Install PyTorch (CPU-only for broader compatibility in Docker) and torchaudio
+# Then install whisperx
+RUN pip install --no-cache-dir \
+    torch==2.2.2+cpu \
+    torchaudio==2.2.2+cpu \
+    -f https://download.pytorch.org/whl/torch_stable.html \
+    && pip install --no-cache-dir whisperx==3.1.1
+
+# Set environment variable to enable demo features
+ENV DEMO_AVAILABLE=1
+
+# Expose the port that Gunicorn will listen on
+EXPOSE 8000
+
+# Run gunicorn to serve the Flask application
+CMD ["gunicorn", "--bind", "0.0.0.0:8000", "app:app"]
+
+# --- Stage 3: Build without WhisperX support ---
+FROM base AS without_whisperx
+
+# Set environment variable to disable demo features
+ENV DEMO_AVAILABLE=0
+
 # Expose the port that Gunicorn will listen on
 EXPOSE 8000
 

app.py

@@ -1,7 +1,7 @@
 from flask import Flask, render_template, request, jsonify, send_from_directory
 from generate_podcast import generate, PODCAST_SCRIPT, setup_logging, validate_speakers, update_elevenlabs_quota
 from utils import sanitize_text, get_asset_path, get_app_data_dir
-from config import AVAILABLE_VOICES, DEFAULT_APP_SETTINGS
+from config import AVAILABLE_VOICES, DEFAULT_APP_SETTINGS, DEMO_AVAILABLE
 from create_demo import create_html_demo_whisperx
 import os
 import tempfile
@@ -55,7 +55,7 @@ def save_settings(settings):
 # --- Routes ---
 @app.route('/')
 def index():
-    return render_template('index.html', default_script=PODCAST_SCRIPT)
+    return render_template('index.html', default_script=PODCAST_SCRIPT, demo_available=DEMO_AVAILABLE)
 
 @app.route('/assets/<path:filename>')
 def get_asset(filename):
@@ -189,6 +189,10 @@ def handle_generate():
 
 @app.route('/api/generate_demo', methods=['POST'])
 def handle_generate_demo():
+    # If DEMO_AVAILABLE is not set to "1", return an error
+    if not DEMO_AVAILABLE:
+        return jsonify({'error': 'Demo generation is not available.'}), 403
+
     data = request.json
     script_text, audio_filename = data.get('script'), data.get('audio_filename')
     title, subtitle = data.get('title', 'Podcast Demo'), data.get('subtitle', '')

config.py

@@ -1,3 +1,5 @@
+import os
+
 SERVICE_CONFIG = {
     "elevenlabs": {
         "title": "ElevenLabs API Key",
@@ -54,4 +56,7 @@
         "Samantha": {"id": "cgSgspJ2msm6clMCkdW9", "display_name": "Jessica - Female, Young, american"}
     },
     "elevenlabs_quota_cache": None
-}
+}
+
+# Environment variable to control the demo button visibility
+DEMO_AVAILABLE = os.getenv("DEMO_AVAILABLE") == "1"

docker-compose.yml

@@ -9,4 +9,5 @@ services:
     environment:
       - ELEVENLABS_API_KEY=${ELEVENLABS_API_KEY}
       - GEMINI_API_KEY=${GEMINI_API_KEY}
+      - DEMO_AVAILABLE=${DEMO_AVAILABLE}
     restart: always

docker_push.sh

@@ -0,0 +1,88 @@
+#!/bin/bash
+
+# This script automates the process of building and pushing two versions of a Docker image
+# to Docker Hub: a full version ('latest') and a lightweight version ('light').
+# Each version is tagged with a specific version, a major version, and a fixed tag.
+#
+# Usage:
+#   1. Make the script executable: chmod +x docker_push.sh
+#   2. Run the script with a version number: ./docker_push.sh <version>
+#
+# Example:
+#   ./docker_push.sh 2.0.0b8
+
+# Exit immediately if a command exits with a non-zero status.
+set -e
+
+# --- Configuration ---
+# Your Docker Hub username
+USERNAME="gandulf78"
+# The name of the image
+IMAGE_NAME="podcast_generator"
+
+# --- Script Logic ---
+
+# Check if a version tag was provided as an argument
+if [ -z "$1" ]; then
+  echo "Error: No version tag provided." >&2
+  echo "Usage: ./docker_push.sh <version>" >&2
+  echo "Example: ./docker_push.sh 2.0.0b8" >&2
+  exit 1
+fi
+
+VERSION=$1
+# Extract the major version number (e.g., "2" from "2.0.0b8")
+MAJOR_VERSION=$(echo "$VERSION" | cut -d. -f1)
+
+# --- Full Version (with WhisperX) ---
+TAG_SPECIFIC_FULL="$USERNAME/$IMAGE_NAME:$VERSION"
+TAG_MAJOR_FULL="$USERNAME/$IMAGE_NAME:$MAJOR_VERSION"
+TAG_LATEST_FULL="$USERNAME/$IMAGE_NAME:latest"
+
+echo "--- Building and tagging FULL Docker image (with WhisperX) ---"
+echo "  > Specific tag: $TAG_SPECIFIC_FULL"
+echo "  > Major tag:    $TAG_MAJOR_FULL"
+echo "  > Latest tag:   $TAG_LATEST_FULL"
+
+# 1. Build the full image using the 'with_whisperx' target
+docker build --target with_whisperx -t "$TAG_SPECIFIC_FULL" .
+
+# 2. Add the other tags to the same image
+docker tag "$TAG_SPECIFIC_FULL" "$TAG_MAJOR_FULL"
+docker tag "$TAG_SPECIFIC_FULL" "$TAG_LATEST_FULL"
+
+echo "--- Pushing all FULL tags to Docker Hub ---"
+# 3. Push all full tags to Docker Hub
+docker push "$TAG_SPECIFIC_FULL"
+docker push "$TAG_MAJOR_FULL"
+docker push "$TAG_LATEST_FULL"
+
+echo "✅ Successfully pushed all FULL tags to Docker Hub!"
+
+# --- Light Version (without WhisperX) ---
+TAG_SPECIFIC_LIGHT="$USERNAME/$IMAGE_NAME:${VERSION}-light"
+TAG_MAJOR_LIGHT="$USERNAME/$IMAGE_NAME:${MAJOR_VERSION}-light"
+TAG_FIXED_LIGHT="$USERNAME/$IMAGE_NAME:light"
+
+echo ""
+echo "--- Building and tagging LIGHT Docker image (without WhisperX) ---"
+echo "  > Specific tag: $TAG_SPECIFIC_LIGHT"
+echo "  > Major tag:    $TAG_MAJOR_LIGHT"
+echo "  > Fixed tag:    $TAG_FIXED_LIGHT"
+
+# 1. Build the light image using the 'without_whisperx' target
+docker build --target without_whisperx -t "$TAG_SPECIFIC_LIGHT" .
+
+# 2. Add the other tags to the same image
+docker tag "$TAG_SPECIFIC_LIGHT" "$TAG_MAJOR_LIGHT"
+docker tag "$TAG_SPECIFIC_LIGHT" "$TAG_FIXED_LIGHT"
+
+echo "--- Pushing all LIGHT tags to Docker Hub ---"
+# 3. Push all light tags to Docker Hub
+docker push "$TAG_SPECIFIC_LIGHT"
+docker push "$TAG_MAJOR_LIGHT"
+docker push "$TAG_FIXED_LIGHT"
+
+echo "✅ Successfully pushed all LIGHT tags to Docker Hub!"
+echo ""
+echo "🚀 All versions have been successfully built and pushed."

gui.py

@@ -40,7 +40,7 @@
 from generate_podcast import validate_speakers, update_elevenlabs_quota
 from utils import get_asset_path, sanitize_app_settings_for_backend, find_ffplay_path, get_app_data_dir, sanitize_text
 from create_demo import create_html_demo_whisperx
-from config import AVAILABLE_VOICES, DEFAULT_APP_SETTINGS
+from config import AVAILABLE_VOICES, DEFAULT_APP_SETTINGS, DEMO_AVAILABLE
 
 # --- Versioning ---
 # This file is automatically generated by setuptools-scm
@@ -92,6 +92,7 @@ def __init__(self, root: tk.Tk, generate_func, logger, api_key: str, default_scr
         self.ffplay_path = find_ffplay_path()
         self.is_whisperx_available = self.check_whisperx_availability()
         self.elevenlabs_quota_text = None  # New state variable
+        self.demo_available = DEMO_AVAILABLE # Read environment variable
 
         self.app_settings = self.load_settings()
         self.provider_var = tk.StringVar(value=self.app_settings.get("tts_provider", "elevenlabs").lower())
@@ -233,16 +234,17 @@ def _setup_ctk_menu(self):
             self.settings_menu.add_option("Quit", command=self.root.quit)
 
         # Menu Actions
-        actions_button = self.menu_bar.add_cascade("Actions")
-        self.actions_menu = CTkMenuBar.CustomDropdownMenu(
-            widget=actions_button,
-            corner_radius=3,
-            border_width=1,
-            border_color=("gray70", "gray25")
-        )
-        self.demo_menu_item = self.actions_menu.add_option("Generate HTML Demo...",
-                                                           command=self.start_demo_generation_thread,
-                                                           state='disabled')
+        if self.demo_available:
+            actions_button = self.menu_bar.add_cascade("Actions")
+            self.actions_menu = CTkMenuBar.CustomDropdownMenu(
+                widget=actions_button,
+                corner_radius=3,
+                border_width=1,
+                border_color=("gray70", "gray25")
+            )
+            self.demo_menu_item = self.actions_menu.add_option("Generate HTML Demo...",
+                                                               command=self.start_demo_generation_thread,
+                                                               state='disabled')
 
         # Menu Help
         help_button = self.menu_bar.add_cascade("Help")
@@ -302,12 +304,13 @@ def _setup_tkinter_menu(self):
             self.settings_menu.add_command(label="Quit", command=self.root.quit)
 
         # --- Actions Menu ---
-        self.actions_menu = tk.Menu(self.menubar, tearoff=0)
-        self._apply_menu_theme(self.actions_menu)
-        self.menubar.add_cascade(label="Actions", menu=self.actions_menu)
-        self.actions_menu.add_command(label="Generate HTML Demo...",
-                                      command=self.start_demo_generation_thread,
-                                      state='disabled')
+        if self.demo_available:
+            self.actions_menu = tk.Menu(self.menubar, tearoff=0)
+            self._apply_menu_theme(self.actions_menu)
+            self.menubar.add_cascade(label="Actions", menu=self.actions_menu)
+            self.actions_menu.add_command(label="Generate HTML Demo...",
+                                          command=self.start_demo_generation_thread,
+                                          state='disabled')
 
         # Help Menu (common to all platforms)
         self.help_menu = tk.Menu(self.menubar, tearoff=0)
@@ -373,6 +376,8 @@ def update_voice_settings_enabled_ctk(self):
 
     def update_demo_menu_state_ctk(self, enabled):
         """Met à jour l'état du menu demo avec CTkMenuBarPlus."""
+        if not self.demo_available:
+            return
         try:
             if hasattr(self.actions_menu, 'configure_option') and self.demo_menu_item:
                 state = 'normal' if enabled else 'disabled'
@@ -885,20 +890,6 @@ def _wait_for_cache():
             # If not using ElevenLabs or cache is ready, open immediately.
             self._show_settings_window()
 
-    def _show_settings_window(self):
-        """Creates and displays the actual settings window."""
-        from settings_window import VoiceSettingsWindow
-        VoiceSettingsWindow(
-            self.root,
-            current_settings=self.app_settings,
-            save_callback=self.save_settings,
-            close_callback=self.on_settings_window_close,
-            default_settings=DEFAULT_APP_SETTINGS,
-            preloaded_elevenlabs_voices=self.elevenlabs_voices_cache,
-            play_gemini_sample_callback=self.play_gemini_voice_sample,
-            play_elevenlabs_sample_callback=self.play_elevenlabs_voice_sample
-        )
-
     def show_about_window(self):
         """Displays the 'About' window."""
         AboutWindow(self.root, version=get_app_version())
@@ -1043,7 +1034,8 @@ def start_generation_thread(self):
             except:
                 pass
         else:
-            self.actions_menu.entryconfig("Generate HTML Demo...", state='disabled')
+            if self.demo_available:
+                self.actions_menu.entryconfig("Generate HTML Demo...", state='disabled')
             self.menubar.entryconfig("Settings", state="disabled")
 
         # Show and start the progress bar
@@ -1125,14 +1117,19 @@ def on_generation_complete(self, success: bool):
         if HAS_CTK_MENUBAR:
             self.update_demo_menu_state_ctk(can_generate_demo)
         else:
-            self.actions_menu.entryconfig("Generate HTML Demo...", state='normal' if can_generate_demo else 'disabled')
+            if self.demo_available:
+                self.actions_menu.entryconfig("Generate HTML Demo...", state='normal' if can_generate_demo else 'disabled')
+
 
         if self.progress_bar.winfo_ismapped():
             self.progress_bar.pack_forget()
         self.log_text.configure(state='disabled')  # Disable the log area at the very end
 
     def start_demo_generation_thread(self):
         """Opens a dialog to get demo settings, then starts the generation."""
+        if not self.demo_available:
+            return
+
         if not self.last_generated_filepath or not self.last_generated_script:
             messagebox.showwarning("No Data", "Please generate a podcast first before creating a demo.",
                                    parent=self.root)
@@ -1156,7 +1153,8 @@ def _on_demo_settings_confirmed(self, title: str, subtitle: str, output_dir: str
         if HAS_CTK_MENUBAR:
             self.update_demo_menu_state_ctk(False)
         else:
-            self.actions_menu.entryconfig("Generate HTML Demo...", state='disabled')
+            if self.demo_available:
+                self.actions_menu.entryconfig("Generate HTML Demo...", state='disabled')
 
         thread = threading.Thread(
             target=self.run_demo_generation,
@@ -1196,8 +1194,9 @@ def run_demo_generation(self, script_content: str, audio_filepath: str, title: s
                 if HAS_CTK_MENUBAR:
                     self.root.after(0, lambda: self.update_demo_menu_state_ctk(can_generate_demo))
                 else:
-                    self.root.after(0, lambda: self.actions_menu.entryconfig("Generate HTML Demo...",
-                                                                             state='normal' if can_generate_demo else 'disabled'))
+                    if self.demo_available:
+                        self.root.after(0, lambda: self.actions_menu.entryconfig("Generate HTML Demo...",
+                                                                                 state='normal' if can_generate_demo else 'disabled'))
             # Clean up the temporary file
             if temp_script_file and os.path.exists(temp_script_file):
                 os.remove(temp_script_file)
@@ -1526,8 +1525,6 @@ def _run_fetch():
                 self.logger.warning(f"Failed to pre-fetch ElevenLabs voices: {e}")
                 self.elevenlabs_voices_cache = []
 
-        threading.Thread(target=_run_fetch, daemon=True).start()
-
 
 def show_error_and_log(*args):
     """Global exception handler with improved debugging."""

requirements.txt

@@ -1,7 +1,13 @@
+# Core dependencies for the Podcast Generator
+
+# Web server
 Flask
 gunicorn
+python-dotenv
+
+# Google and ElevenLabs APIs
 google-genai
 elevenlabs
+
+# Utility
 requests
-python-dotenv
-keyring