Translate first, then dub one foreign cut | AI Video Factory

Overseas views need more than captions. If the platform can play a foreign voice, stacked translation alone feels like homework. In the studio the calmer path is: review the source cues, translate on the same timeline, then dub from the translation. Timestamps never move. Dubbing voices are Chinese 晓晴 and English; Japanese, Korean, German, French and Russian can be caption languages — spoken takes stay Chinese or English.

Translate the words, then speak

A dirty source axis scales mistakes in both translation and speech. Do not reverse the order:

1. Build timestamped source captions (speech, burned-in OCR, or an imported SRT).
2. Fix typos, names and breaths, then start translation.
3. Review names and line length in the translation, then start dubbing.
4. If you need burned-in type, mix the audio first, then burn. Do not burn and then dub.

The ten-minute caption path is in this guide. Bilingual captions without a new voice: translation.

A foreign cut on the same timeline

1. Open translation in the studio with the current project or a local MP4 / MOV. The file stays in the browser.
2. Confirm the source cues are reviewed. Tick caption translation. Pick English first when you want an English voice. One target language per run.
3. Scan names and long lines. On 9:16, split past ~14 Chinese characters or a short English clause — see vertical layout.
4. Open dubbing. English picks from several voices; Chinese back-translation uses 晓晴. Auto-detect splits Chinese and English lines.
5. For a spoken foreign cut, turn original audio off so two languages do not stack. The picture stays locked to the source.
6. Export the mixed soundtrack first, then burn the translated captions, or save that language as SRT / VTT.

The free plan includes one translation run and one dubbing run per month. Languages, voices and limits are on Features and Pricing.

Check only three things

• Names: match brands and model numbers to the source before the voice reads them.
• One breath per cue: when the translation grows, split the line or change in/out — do not drag the whole file’s speed.
• Original audio: off for a spoken foreign cut; leave it on if you only stack bilingual captions and keep the Chinese voice.

What usually breaks

Dub Chinese first, then translate: the voice still reads the source and the new lines miss the mouth. Translate, then dub. Caption-only Japanese, Korean, German, French or Russian has no matching voice engine — do not force an English voice to read those languages. Keep the Chinese original or cut a separate English version. Export a cover in the target ratio — see covers from the clip.