How to Auto Generate Subtitles in Kdenlive

This guide provides a step-by-step walkthrough on how to use Kdenlive’s built-in speech-to-text feature to automatically generate subtitles for your video projects. You will learn how to install the required speech recognition engine, download language models, and automatically translate your video’s audio track into clean, synced on-screen text.

Step 1: Install the Speech Recognition Engine (Vosk)

Kdenlive utilizes Python and the Vosk speech recognition toolkit to convert audio to text.

  1. Open Kdenlive and navigate to Settings > Configure Kdenlive (on macOS, Kdenlive > Preferences).
  2. Select the Speech to Text tab from the left sidebar.
  3. Check if Python is installed. If not, Kdenlive will prompt you to install it.
  4. Click the Install Vosk button to download and set up the speech recognition library.

Step 2: Download a Language Model

Once Vosk is installed, you need a language model that matches the audio in your video.

  1. In the same Speech to Text settings window, click on the link provided to view available Vosk models, or click the Check / Download button next to the model list.
  2. Select your desired language (e.g., English, Spanish, French). For general use, the “small” models are recommended as they are faster to download and process, while the larger models offer higher accuracy at the cost of speed.
  3. Once downloaded, Kdenlive will extract the model automatically. Ensure the correct model is selected in your Speech to Text settings, then click Apply and OK.

Step 3: Open the Speech Recognition Tool

With the software configured, you can now generate subtitles for your active timeline.

  1. Ensure your video and audio clips are placed on the timeline.
  2. In the top menu, go to Project > Subtitles > Speech Recognition.
  3. If the Subtitle track is not already visible, Kdenlive will automatically enable it at the top of your timeline.

Step 4: Configure and Run the Speech-to-Text Process

A dialog box will appear with options for generating your subtitles.

  1. Model: Select the language model you downloaded in Step 2.
  2. Apply to: Choose whether to analyze the Timeline (entire project) or just a Selected Clip.
  3. Device: Keep this on “Default” unless you have specific audio routing needs.
  4. Click the Process button.

Kdenlive will analyze the audio track. A progress bar will show the transcription status.

Step 5: Edit and Refine the Subtitles

Once processing is complete, the generated subtitles will appear as blocks on the Subtitle track directly above your video tracks.

  1. To edit a subtitle, click on the subtitle block in the timeline.
  2. Edit the text in the Subtitle monitor panel on the right side of the screen to correct any spelling or timing errors.
  3. To adjust the duration of a subtitle, drag the edges of the subtitle block left or right on the timeline.

Step 6: Export Your Subtitles

You can either render the subtitles directly into your video file (hardcoded) or export them as a separate file.