Minor title change

Author: meizhong986

notebook/WhisperJAV_colab_parallel_expert.ipynb

@@ -4,7 +4,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "id": "cell-0",
-   "source": "<a href=\"https://colab.research.google.com/github/meizhong986/WhisperJAV/blob/main/notebook/WhisperJAV_colab_parallel_expert.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>\n\n# WhisperJAV Two-Pass Edition v1.7.4 (Expert)\n\n**Adaptive Two-Pass Processing** - Automatically optimizes for your platform.\n\n| Platform | GPUs | How it works |\n|----------|------|---------------|\n| **Kaggle** | 2x T4 (8GB each) | **Parallel** - Pass 1 on GPU 0, Pass 2 on GPU 1 simultaneously |\n| **Colab L4/A100** | 1x GPU (16-24GB) | **Sequential** - Pass 1 first, then Pass 2 (avoids memory issues) |\n\n| Option | What it controls |\n|--------|------------------|\n| **Scene Detection** | How to split audio into chunks (auditok, silero, semantic) |\n| **Speech Segmenter** | How to detect speech in audio (silero, ten) |\n| **Speech Enhancer** | Audio cleanup for noisy sources (ffmpeg-dsp, clearvoice, etc.) |\n| **Model** | Which AI model to use (large-v2, large-v3, turbo, kotoba) |\n\n---\n<div style=\"font-size: 8px; line-height: 1.0;\">\n1. Upload your videos to <code>Google Drive/WhisperJAV/</code><br>\n2. Run <b>Step 1: Settings</b> (required)<br>\n3. Run <b>Step 1.5: Expert Options</b> (optional - skip if unsure)<br>\n4. Run <b>Step 2: Two-Pass Transcribe</b> and wait for completion<br>\n5. Run <b>Step 3: AI Translation</b> (if selected)\n</div>\n\n<small>The notebook will automatically disconnect when finished to save your GPU credits.</small>"
+   "source": "<a href=\"https://colab.research.google.com/github/meizhong986/WhisperJAV/blob/main/notebook/WhisperJAV_colab_parallel_expert.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>\n\n# WhisperJAV (parallel) Two-Pass Edition v1.7.4 (Expert)\n\n**Adaptive Two-Pass Processing** - Automatically optimizes for your platform.\n\n| Platform | GPUs | How it works |\n|----------|------|---------------|\n| **Kaggle** | 2x T4 (8GB each) | **Parallel** - Pass 1 on GPU 0, Pass 2 on GPU 1 simultaneously |\n| **Colab L4/A100** | 1x GPU (16-24GB) | **Sequential** - Pass 1 first, then Pass 2 (avoids memory issues) |\n\n| Option | What it controls |\n|--------|------------------|\n| **Scene Detection** | How to split audio into chunks (auditok, silero, semantic) |\n| **Speech Segmenter** | How to detect speech in audio (silero, ten) |\n| **Speech Enhancer** | Audio cleanup for noisy sources (ffmpeg-dsp, clearvoice, etc.) |\n| **Model** | Which AI model to use (large-v2, large-v3, turbo, kotoba) |\n\n---\n<div style=\"font-size: 8px; line-height: 1.0;\">\n1. Upload your videos to <code>Google Drive/WhisperJAV/</code><br>\n2. Run <b>Step 1: Settings</b> (required)<br>\n3. Run <b>Step 1.5: Expert Options</b> (optional - skip if unsure)<br>\n4. Run <b>Step 2: Two-Pass Transcribe</b> and wait for completion<br>\n5. Run <b>Step 3: AI Translation</b> (if selected)\n</div>\n\n<small>The notebook will automatically disconnect when finished to save your GPU credits.</small>"
   },
   {
    "cell_type": "code",
@@ -174,4 +174,4 @@
  },
  "nbformat": 4,
  "nbformat_minor": 5
-}
+}