Kdenlive Adds AI Speech-to-Text Subtitle Generation in Latest Release

Kdenlive has released a major update introducing automatic subtitle generation powered by local AI speech recognition, making it the first major open-source video editor to include built-in transcription without requiring a cloud service. The feature uses OpenAI's Whisper model running locally on the user's machine to transcribe audio from video clips and automatically generate subtitle tracks. The transcription runs entirely offline, with no audio data sent to any external server. Users can select from multiple Whisper model sizes depending on their hardware — smaller models run faster on modest hardware, while larger models provide better accuracy for complex audio. The generated subtitles can be edited directly in the timeline and exported in SRT, VTT, or ASS formats. The update also includes improved color scopes for professional color grading, a new audio waveform display in the timeline, and faster rendering for H.265 and AV1 output. Kdenlive is available as a free download for Windows, macOS, and Linux under the GPL-2.0 license.