An automated video globalization tool that transcribes, translates, and re-dubs videos using AI. 一个自动化的视频全球化工具，利用 AI 实现视频的转录、翻译和自动配音。

• Auto Transcription: Uses OpenAI Whisper to detect and transcribe original audio with precise timestamps.
• 智能转录：使用 OpenAI Whisper 自动检测并转录原始音频，带有精确的时间戳。
• Multilingual Translation: Seamlessly translates content into Chinese, Spanish, French, and more.
• 多语言翻译：无缝将内容翻译为中文、西班牙语、法语等。
• Professional Dubbing: Integrated with edge-tts for high-quality, natural AI voices.
• 专业配音：集成 edge-tts，提供高质量、自然的 AI 语音。
• Auto-Sync & Speed Correction: Automatically adjusts voiceover speed to match video timing and prevent overlaps.
• 自动同步与语速校正：自动调整配音语速以匹配视频时长，防止语音重叠。
• Dynamic Captions: Renders cinematic captions directly into the final video.
• 动态字幕：将电影级字幕直接渲染到最终视频中。

• Frontend: Next.js (App Router), TypeScript, Vanilla CSS
• Backend: Next.js API Routes, Python Scripts
• Transcription: OpenAI Whisper
• Translation: Deep-Translator (Google)
• Voiceover: Microsoft Edge TTS
• Video Rendering: Remotion, FFmpeg

Ensure you have the following installed: 确保您已安装以下工具：

• Node.js (v18 or later)
• Python 3.9+
• FFmpeg (Required for audio/video processing)
• Pip packages:

  pip install openai-whisper edge-tts deep-translator

1. Clone the repository | 克隆仓库:

   git clone <your-repo-url>
   cd cinematic/translator-app

2. Install Node dependencies | 安装 Node 依赖:

   npm install
   cd remotion && npm install && cd ..

3. Start the development server | 启动开发服务器:

   npm run dev

4. Start Remotion Studio (Optional) | 启动编辑器 (可选): If you want to live-edit the results: 如果您想实时编辑结果：

   npm run studio

5. Access the App | 访问应用: Open http://localhost:3000 in your browser. The editor will be available at http://localhost:3001.

1. Upload Video | 上传视频: Select an MP4 or MOV file from your computer.
2. Select Language | 选择语言: Choose your desired target language (e.g., Mandarin Chinese).
3. Wait for Analysis | 等待分析: The system will run Whisper to transcribe and translate.
4. Choose a Voice | 选择配音:
   • Browse the list of available AI voices.
   • Click "Play Sample" to hear a preview of the voice.
5. Render | 渲染: Click "Render Final Video". The system will generate voiceovers and composite the final video.
6. Download | 下载: Once complete, watch the preview and click "Download Video".

• /src/app: Next.js frontend and API routes.
• /scripts: Core Python logic for Whisper, Translation, and Audio processing.
• /remotion: The video composition engine and assets.
• /public/uploads: Temporary storage for uploaded videos.
• /public/rendered: Final processed video outputs.

• First Run: The first transcription might take longer as Whisper downloads its model (default: base).
• 首次运行：第一次转录可能需要较长时间，因为 Whisper 需要下载模型（默认为 base）。
• Performance: Rendering speed depends on your CPU/GPU and the length of the video.
• 性能：渲染速度取决于您的 CPU/GPU 以及视频长度。

MIT