Choose Your Task

Install GeekLink, choose OCR, Transcribe, or Translate, and complete your first subtitle task.

System Requirements

Item Minimum Requirement
macOS macOS 13 Ventura or later; Apple Silicon (M1/M2/M3/M4) only. Intel is not supported.
Windows 64-bit Windows desktop
Memory At least 4 GB (8 GB or more recommended for better recognition performance)
Disk Space Allow several GB if you download larger speech-recognition models

Installation

  1. Download the installer Go to geeklink.dev and download the macOS DMG or Windows ZIP for your computer.
  2. Install or extract GeekLink On macOS, open the DMG and drag GeekLink to Applications. On Windows, extract the ZIP before launching GeekLink.
  3. Launch for the first time Open GeekLink from Applications on macOS, or from the extracted folder on Windows.

Choose Your Task

GeekLink now opens with one question: What do you want to do today? Choose the workspace that matches your job:

After you choose, GeekLink keeps that workspace focused. The small Change task link returns to the chooser when you need a different job.

GeekLink's three-task home screen with OCR, speech recognition, and subtitle translation
Choose the job you came for. Each card opens a focused workspace.

Try a Built-In Demo

Each empty workspace shows only the demo relevant to that task. OCR loads an OCR clip; Transcribe loads a speech clip. Translation deliberately loads its demo in two steps — first the video, then its matching source SRT — so you can see that translation needs both.

GeekLink OCR workspace with video-only import and the OCR Library
OCR keeps the workspace focused on extracting hardcoded subtitles from video.
GeekLink Transcribe workspace with speech language, collapsed advanced settings, and Watch folder
Transcribe keeps advanced recognition settings collapsed until you need them.
GeekLink Translation workspace with a demo video and its source subtitle track loaded
Translation becomes ready after the demo video and its source subtitles are both loaded.

Run Your First Task

Here is the speech-to-translation flow when you do not already have a source SRT:

  1. Import a video Choose Transcribe speech, then drag in a video or load the Speech Recognition Demo.
  2. Configure recognition Choose the source language and recognition model. Expand More settings only when you need the advanced options.
  3. Run Speech Recognition GeekLink creates editable source subtitles from the spoken audio.
  4. Continue to Translate When recognition finishes, use Continue to Translate. GeekLink opens the Translation workspace with the recognized source subtitles ready.
  5. View the results Choose the target language and engine, run translation, then open the subtitle editor to review the source and translated tracks.
Tip The first time you use a new recognition model, the system will automatically download the model file. Please be patient. Once the download is complete, subsequent uses will start recognition immediately.

Next Steps

Once you know which workflow fits, use the focused guide for that product: