The short answer
Use subtitles when viewers need to follow the message without sound or when keeping the original voice matters. Consider dubbing when listening should carry the explanation while the viewer watches the product. For a demo that needs to work in both situations, prepare a reviewed dubbed version and matching captions. Neither choice fixes inaccurate product claims or untranslated text inside the picture.
What changes with dubbing and with subtitles?
Dubbing creates a new spoken-language version of a video. Subtitling presents timed text while leaving the spoken track available. They solve different parts of the same problem: a viewer needs to understand the product while following what happens on screen.
This guide is about short product demonstrations, explainers, and introductions—not choosing how to watch a foreign-language film. A product video has concrete details to protect: a model name, a feature demonstration, a dimension, a quantity, or an instruction. An elegant translation is not enough if it changes one of those details.
We use captions to mean text that also conveys important non-speech audio, and translated subtitles for text in another language. Terminology varies by region. Dubbing alone does not provide a text alternative for people who cannot hear the audio.
Choose around the viewing situation
Start with where the video will actually play. Do not choose a method just because a country is assumed to prefer it. Watch the proposed version on the destination device, with sound both on and off.
| Situation | Starting choice | Check before release |
|---|---|---|
| The message must work with sound off | Captions or subtitles | Can a viewer understand the offer and action without hearing anything? |
| A viewer needs to watch a detailed demonstration | Dubbing plus optional captions | Does reading compete with the small parts or controls being shown? |
| The original speaker’s delivery is central | Original audio plus translated subtitles | Is the text readable without losing the speaker’s emphasis? |
| The video will serve sound-on and sound-off placements | Dubbed audio and reviewed captions | Do the two versions communicate the same product facts? |
| The visible mouth must match the translated words | A workflow explicitly supporting lip-sync | Ordinary dubbing is not proof of lip-sync support. |
When a subtitle-first version makes sense
Try subtitles first when the original audio should stay intact and the frame has room for readable text. For a simple demonstration with short sentences, a translated subtitle track can carry the explanation without replacing the performance.
Decide between a separate track and burned-in subtitles. A separate file works only when the destination player supports and exposes it. Burned-in text becomes part of the image and cannot be switched off. Keep a clean master so future translations or corrections do not have to cover old text.
Check the actual mobile frame. A caption may be grammatically correct but cover the feature being demonstrated. Shorten the wording without removing a warning, number, or product qualification. If the information cannot fit, change the edit or narration rather than simply shrinking the type.
When a dubbed version is worth testing
Consider dubbing when the explanation is easier to follow by listening while looking at the product. A close-up of a latch, an assembly step, or a moving interface can leave little spare attention for reading. A new voice track is useful only if its timing still points to the correct action.
Listen without reading the transcript first. Can you recognize the product name? Are measurements clear? Does a sentence finish before the next step begins? Then compare the words against the approved product facts. A natural-sounding sentence can still be wrong.
Revisions also matter. A subtitle correction requires a text-and-timing check; a spoken correction may require new audio and another video review. Budget for review and rework, not just the first export. Do not assume one method is always cheaper without comparing the actual scope.
A product-demo decision, without invented results
Imagine a short travel-mug demonstration. The presenter says “Keep the lid closed” while showing the seal. This is an illustrative scenario, not a customer case study or a measured performance result.
For a silent placement, those words need to be readable while the seal is visible. For a spoken walkthrough, the translated instruction should arrive at that same moment. If captions cover the seal, move or shorten them in an editor. If the voice instruction comes after the shot changes, fix the timing before release.
If you test both versions with real viewers, keep the offer, destination and audience as consistent as possible. Define the outcome you are comparing in advance. A few favorable comments are useful feedback, but not evidence of a conversion-rate increase.
What this means for a Recast workflow
Recast Studio accepts Chinese or English source speech and supports English, German, French, Japanese, and Spanish targets, with up to four targets in one job. Its 12 website interface languages are a separate setting. Check the current feature and upload limits before preparing a source.
Studio produces translated speech with MP4 and SRT downloads. The presenter and product stay in the source footage; framing follows the selected output ratio. Recast does not provide presenter replacement or lip-sync, and it does not automatically translate text already visible inside the image.
The free subtitle converter changes file formats, not languages. The free subtitle burner uses an existing subtitle file; it does not translate or transcribe speech. Review every localized version before publishing, including speaker continuity when more than one person speaks.