AI Voiceover Translation for Video

SubtitleKit's AI voiceover translation runs entirely in your browser. Drop a video up to 60 minutes long and it finds or transcribes the dialogue, translates it, and generates a natural AI narrator voice in the language you choose. The original audio stays underneath at a lower volume by default, or you can replace it entirely. Free AI dubbing, with no sign-up and no watermark on the export.

How AI voiceover translation works

The tool looks for subtitles already embedded in your video, or a subtitle file you drop alongside it; if neither exists, a Whisper AI model transcribes the speech. The dialogue is translated into your target language, then an AI voice model reads the translation aloud, one line at a time, timed to fit the original speech. Transcription, translation, and voice synthesis all run in the browser tab, which is why a 60-minute video costs the same as a 60-second one.

How to use this tool

• Drop your video. The tool automatically finds existing subtitles, or transcribes the speech with AI if there aren't any.
• Pick a language and voice. Choose the language to translate into and one of 10 AI narrator voices, then preview it on a real line.
• Generate the voiceover. The first run needs a one-time 383 MB AI voice model download. Your browser then translates each line and synthesizes the narration.
• Download. Choose whether to lower the original audio under the narration or replace it entirely, then download your video with the new voiceover.

What is AI voiceover translation?

AI voiceover translation, often called AI dubbing, replaces or layers a new spoken-language track over a video's original audio, generated automatically from a translated transcript instead of by a voice actor. SubtitleKit's version is a single AI narrator reading the translated lines (a "lektor" style, common for documentaries and tutorials), not a full multi-voice production. Most AI dubbing tools charge per minute of video or stamp a watermark across the export; here the transcription, translation, and voice synthesis all happen inside your browser, so neither applies.

What this tool needs

A video file is all you need. The tool figures out the rest.

Video files

• MP4, MOV, MKV, and WebM, up to 60 minutes long.
• Embedded subtitle tracks are detected and reused automatically when present.
• No subtitles yet? A Whisper AI model transcribes the speech first.
• 31 voiceover languages, spanning Latin, Cyrillic, Greek, Arabic, Devanagari, and CJK scripts, with deep coverage across Europe alongside Arabic, Hindi, Indonesian, Vietnamese, Japanese, and Korean.

What you get

• The same video, with a new AI-narrated audio track in your chosen language.
• Choose to lower the original audio underneath the narration, or replace it entirely.

Why this voiceover translator is different

Cloud dubbing tools charge per minute of video, watermark the free tier, and queue your export. SubtitleKit translates and generates the voiceover in the browser tab instead.

• Smart subtitle reuse. The tool checks for an embedded subtitle track or a dropped subtitle file before ever running speech recognition, so it never does unnecessary work.
• Editable before synthesis. Review and fix the translated lines in the built-in cue table before generating the voiceover.
• Natural voice timing. The AI narrator's pacing is fitted to each line's original timing, speeding up slightly only when a translation runs long.
• Lower or replace. Keep the original audio audible underneath the narration, or replace it entirely, your choice per export.
• Works offline after setup. Once the one-time voice model download completes, voiceover generation needs no internet connection.