How to dub an English video using SRT subtitles

REAL ENGLISH AUDIO

Listen: The museum after closing

A real English dubbing audio example, not a rendered video. Tim voices the guide and Dacey voices the visitor. Download the matching SRT to inspect the cue boundaries. The timestamps were measured from this audio output.

Generated 2026-10-26 · 15.44 seconds · Real speech-service output · No generation request is made when you play this file.

Source text for this example

Guide: Welcome to the night gallery. Please stay behind the silver line.
Visitor: Why is the painting moving? I thought the exhibition was closed.
Guide: It is closed to visitors. The paintings have only just arrived.

These are original example texts. The audio uses the role-labelled version shown above.

1. Upload the video and choose a subtitle source

Open the video studio and create a project. Upload a video you have permission to process. Import an existing SRT file if you have one, or use subtitle recognition and review its result. Recognition may make mistakes; check proper names, punctuation and line order.

The audio example on this page has no source video. Its matching SRT is a practice file for understanding timing and roles. Use your own video footage for an actual dubbing project, and make sure its duration covers the subtitle timeline.

2. Inspect the cue boundaries, not only the words

Each SRT block needs an index, a start and end timestamp, and spoken text. Check for overlapping cues and invalid end times. A subtitle that stays on screen briefly may contain more text than a voice can say naturally in that interval.

Listen to the downloadable audio while following the matching SRT. Its cue lengths come from the generated speech, with 0.45-second gaps. In your own video, the picture determines the available timing. Shorten a line or adjust its delivery when the speech does not fit comfortably.

3. Choose voices for each subtitle line

Use role labels such as Guide: dialogue and Visitor: dialogue to separate speakers. Open a line in the right-hand Subtitle dubbing list to choose its voice, emotion and speed. The left-hand tabs contain Video preview, Edit subtitles and Finished video; there is no separate right-hand role-settings column.

The published example uses Tim for the guide and Dacey for the visitor. Keep the spelling of role labels consistent. Do not put a character name inside the spoken sentence unless you want that name read as dialogue.

4. Generate, preview and export the video

Review the Delete original video audio option in the shared bottom bar. Checked means only the dubbed audio remains; unchecked mixes the original audio at 25%. Enable Generate subtitles only if you want the edited dialogue rendered onto the picture; it does not erase existing text. Generate after reviewing the credit estimate, then use the left Finished video tab to review the result.

Download the MP4 result or WAV, MP3 or FLAC dubbing audio when ready. Source video media is retained for seven days under the current workflow, so download important outputs before expiry. The feature speaks your edited subtitles; it does not automatically translate them or synchronize lips.

Video dubbing: edit SRT, compare subtitles and preview

The left tabs are Video preview, Edit subtitles and Finished video. Edit subtitles contains the SRT editor, Reset subtitles, Clear subtitles and subtitle comparison. Finished video is disabled until a result exists. The right side remains the editable dialogue list.

Overlapping subtitles stay visible in the list and every conflicting timestamp is highlighted red. Correct the start or end times with More actions → Edit timing, or edit the SRT. The red mark clears once the overlap is resolved. Invalid overlaps must be corrected before generation.

In the bottom bar, Delete original video audio leaves only the dubbing when checked; otherwise the source audio is mixed at 25%. Generate subtitles adds the edited dialogue to the picture. It does not erase existing on-screen text, perform OCR cleanup, translate dialogue or synchronize lips.

Finished video downloads support MP4, MOV, MKV and WebM; dubbing audio supports WAV, MP3 and FLAC. The browser preview remains separate from the download format. Choose watermark and format options before downloading and review the shown credit cost; watermarked video downloads currently cost no download credits, while watermark-free downloads use the displayed rate.

Writing prompts for AI narration, dialogue and game voices

Treat these prompts as writing briefs for preparing your script, not commands to paste into spoken dialogue. Use the existing emotion, speed and pause controls for delivery. A role label such as Narrator: or Merchant: identifies the speaker and is not spoken as part of the line.

Narration brief: write a calm 20-second scene in two short sentences, with no stage directions. Dialogue brief: write six alternating lines for two characters with consistent role labels. Game brief: write one greeting, one warning and one farewell for a fantasy merchant; choose an available character preset, then preview each line.

For SRT video dubbing, ask for short, speakable lines that preserve the meaning and names within your existing timestamps. Count timing by listening: shortening the wording is often better than extreme speed. Review recognition errors and overlapping timestamps before generating the full video.

Before you export

• Every subtitle has valid, non-overlapping timestamps.
• The spoken line fits its time range.
• Role labels match the intended voices.
• Original audio does not obscure the new dialogue.

Questions about this workflow