README.md

---
title: Deepgram SRT Generator
emoji: 🎙️
colorFrom: indigo
colorTo: purple
sdk: gradio
sdk_version: 6.18.0
app_file: app.py
pinned: false
license: mit
---

# Deepgram SRT Generation

This project extracts audio from a video file, sends it to Deepgram for transcription, and generates an SRT file with captions.

## Setup

1. Clone the repository:
    ```sh
    git clone https://github.com/bradcypert/deepgram-srt-generation.git
    cd deepgram-srt-generation
    ```

2. Install the required dependencies:
    ```sh
    pip install -r requirements.txt
    ```

3. Set your Deepgram API key as an environment variable:
    ```sh
    export DEEPGRAM_API_KEY=your_deepgram_api_key
    ```

## Usage

Run the script with the path to your video or audio file as an argument:
```sh
python main.py path/to/your/file.mp4
```

This will generate an SRT file with captions in the same directory as your media file.

### CLI Options

You can customize the transcription with the following flags:
*   `-m` / `--model`: Deepgram model to use (default: `nova-3`).
*   `-l` / `--language`: Set the language tag (e.g. `ko`, `en`, `es`).
*   `--no-diarize`: Disable speaker diarization.
*   `-t` / `--translate-to`: Translate the generated subtitles using DeepL.

Example transcribing Korean audio and translating it to English:
```sh
export DEEPL_API_KEY=your_deepl_api_key
python main.py path/to/your/korean_audio.mp3 -l ko -t en
```

### Translate-Only Mode

If you already have an SRT file and want to translate it to another language without transcribing again, simply pass the `.srt` file and target language:
```sh
python main.py path/to/your/subtitles.srt -t ja
```

## Dependencies

- `httpx`
- `moviepy`
- `deepgram`
- `deepgram_captions`
- `deepl`
- `python-dotenv`

Make sure to install these dependencies using `pip` if they are not already installed.

## License

This project is licensed under the MIT License.