The current statistics on video are astounding:

• 82% of internet users watch videos weekly
• About 3.5 billion people around the world watch digital video content,
• People spent 17 hours/week watching videos in 2023

These huge numbers highlight the extensive reach of video as a medium across different demographics and regions.

It’s a logical conclusion that if you are a global company, it’s critical that you make your marketing and product videos available worldwide.

But what does that mean? Dubbing and subtitles are the best places to start. Yet, there’s often confusion between these two terms and a lack of clarity about how they work. It’s time to clarify the difference between these two approaches and explain how to implement them in your video localization program.

OK, let’s start with the basics. What’s the difference between subtitling and dubbing?

Dubbing and subtitling translation services refer to rendering audio dialog into another language inside a video format, such as a movie, TV show, or online video. Subtitling involves placing a written translation of spoken dialogue on screen simultaneously with the audio. (Take a look at our blog post on subtitling for some examples.)

Dubbing, often confused with voiceover, means replacing the spoken dialogue with spoken translation, in most cases simultaneous. Voiceover, on the other hand, is when a narrator provides additional information or commentary without replacing the original audio.

Subtitling vs dubbing: How or why do you choose one over the other?

Sometimes, it’s just a stylistic preference. There’s also a difference in cost that may factor into a decision: subtitling is often less expensive than dubbing because you don’t have to hire native-speaking voice talent.

However, dubbing may be a better solution if your video includes people speaking because subtitles are visible on screen and can be distracting. If there are no visuals of people, then subtitles are less distracting. However, take care that they don’t cover up any on-screen words in the video.

Another factor to keep in mind is that high-profile people who appear on camera often prefer translated subtitling to preserve the integrity of their vocal reputation.

Do people care about “the integrity of their vocal reputation”?

Yes, they do. Sometimes, people don’t want their voices substituted with other voices on recorded media, even if they’re not speaking the same language as the listener, because they want their voices to be identified with them just like they would with their face or body language.

How does subtitling work?

Let’s start with subtitling since it’s a slightly easier process.

First, you will need a transcript of the English dialogue. A transcription service can do this for you.

Now, you (or your dubbing and subtitling translation services vendor) will need to add time codes so that they know exactly when the subtitles are supposed to appear and disappear so they will be aligned with the cadence of speech. Once that’s done, you can start translating.

But be warned: you only have 32 characters on two lines, or 64 characters in total, including spaces. You may need to adjust your subtitles accordingly, which may require slight contextual variations from the original audio dialogue.

But can’t I subtitle things online for free?

Yes. AI can be used to translate and create subtitles that adhere to specific guidelines, such as a 64-character limit per subtitle line. It goes like this:

1. Translation: AI can automatically translate the spoken dialogue into the target language.
2. Segmentation: The text needs to be segmented into subtitle blocks after translation. AI can segment the text into manageable parts while considering natural pauses in speech and ensuring that the meaning remains clear.
3. Adjusting for character limits: AI can be programmed to adhere to specific subtitle standards, including character limits per line. For a 64-character limit, the AI would automatically adjust the text by using synonyms that are shorter, rephrasing sentences, or strategically placing line breaks.
4. Timing: AI can also synchronize subtitles with the audio. It analyzes the audio to determine when a person starts and stops speaking and matches the subtitles to these audio cues, ensuring that they appear and disappear at appropriate times.
5. Review and Adjustment: While AI can handle the bulk of the work efficiently, we always recommend human intervention for final review, especially to ensure cultural appropriateness, local idioms and to catch any nuanced errors the AI might have missed.

But you may find that this process is frustrating. Boiling down the translations into the required number of characters per line while faithfully following the original meaning and audio speed and leaving enough time for the reader to read through it all comfortably can be a challenge.

Although automated subtitling is widely available, it may not your best option if you’re looking for professional, consistent, high-quality results. Bad subtitling with mistranslations and confusing typos can harm your brand’s reputation. Your best bet is to work with a localization agency that leverages qualified, native linguists to ensure that your subtitles accurately represent the original meaning and convey the right tone for your target audience.

Got it. So am I done with subtitling?

Not quite yet. It’s a good idea to have a native speaker review the subtitles and ensure they match the dialogue. Then, pat yourself on the back - you are ready to roll the tape!

Let’s move on to dubbing.

How does dubbing work?

Dubbing, meaning changing the verbal dialogue to a different language, is a bit more complicated. Like before with subtitling, you’ll need to first get your script together and format it with timecodes.

Next, you’ll need to hire native-speaking voice talent to record the translations in a studio. Make sure the translators are aware it’s a voiceover script so they know to stick to the rough length of the original. Often, this recording is done while they are watching the video so they can match their speech to the character speaking on screen.

There are a few different styles of doing this, ranging from a general lowering of the volume to having the character speak free-form until the end of the sentence all the way to lip-syncing in precision with the words spoken by the character.

The latter is more expensive as translating and producing in a studio takes more effort, but it’s a better option for more high-end, polished dubbing. Finally, you’ll need the sound engineers to place the dubbed track into the original audio track and produce the final video.

But can’t I use AI for dubbing?

Yes!

AI-driven dubbing technologies enhance efficiency, reduce costs, and speed up the dubbing process while aiming to maintain the naturalness and emotion of the original performances. Here’s an overview of the main AI options for dubbing in other languages and how they work:

1. Automated Dialogue Replacement (ADR) Tools. AI ADR tools use advanced speech recognition and synthesis technologies to automate lip-syncing. These systems analyze the lip movements in the video and synchronize the dubbed audio to match these movements.
2. Voice Cloning: This technology creates a digital replica of a voice actor’s voice, which can then be used to speak translated text in the target language. This is particularly useful for preserving unique voice qualities.
3. Neural Machine Translation (NMT) and Text-to-Speech Engines: NMT first translates the script. This translation is then adapted for dubbing, considering lip movement synchronization and sentence length to fit the spoken parts. Then, Advanced Text-to-Speech (TTS) engines are employed to deliver natural-sounding speech in the target language.
4. AI-Assisted Dubbing Software: These tools provide automated script translations, suggest lip-sync adjustments, and even recommend voice actors from a database suited for specific roles based on timbre, pitch, and emotional expression.

AI dubbing technologies make the process faster and more cost-effective and open up new possibilities for preserving the emotional and cultural essence of multimedia content across languages.

Does a native speaker need to review dubbing as well?

Yes! Like with subtitles, dubbing needs to be reviewed, and sometimes edits (called “pickups”) must be re-recorded to cover any glitches.

This helps ensure that your dubbed content meets the same quality standards as the original.

What if I have several different languages to dub? Can one person do them all?

Very unlikely. You’ll need to hire individual voice talent for each language into which you want to dub.

Most people, even polyglots, only speak so many languages. And generally speaking, most multilingual people speak some of those languages better than others.

Trying to find professional voice talent who could dub your content with equal language proficiency into, say, six different languages - or even three - would be hunting for unicorns.

You can also use human processes for your top languages and top markets and go with a less expensive automated process for markets that would otherwise not get localized video.

How do I create subtitles and voiceovers that speak to my international target audiences?

To help your brand connect with audiences in new markets, find expert, in-country voice actors and linguists. A translation agency with experience in professional subtitling and dubbing translation solutions can help you find the best talent available for your multilingual voiceover and subtitling work.

Even if you wish to automate much of the process or use synthetic voices, an AI-adept localization partner can help you achieve the quality your business requires.

Where do I learn more about dubbing and subtitling translation services?

To get advice on how best to adapt your video for international markets, including dubbing and/or subtitling, contact one of our multimedia translation experts for more information today.