Atlas 4cwembs21w

HeyGen Help Center

Welcome to Brand Voice! Our new feature for video translation lets you customize translations to match your brand’s identity. With Brand Voice, you can control which words stay the same, which words get a specific translation or pronunciation, and even set the tone of your translations.

: Ensures that key terms remain consistent across all translations.

: Allows for specific customization to align with the brand's identity and tone.

: Helps correct common translation errors and maintain the intended meaning of terms.

You can create and edit multiple brand voices for different video translation projects based on specific translation needs. 

When in Video Translation, once you click on creating a new brand voice, you can start adding your specific needs and preferences for translations under these four areas:

This area is to insert words you don't want translated at all. This is particularly useful for brand names, product names, or technical terms that should remain consistent across different languages. 

If your brand name is Flower Medical, you wouldn't want that to translate into the actual words Flower and Medical, so you can simply insert those words here to not be translated. 

If you're translating a video with technical details like "neural network component" which you prefer to keep in the original language (English), this would be the place to insert it.

This feature allows users to enforce specific translations for certain words to correct any subtle inconsistencies that may arise due to language-specific nuances. 

You can set it up so every time the words "artificial intelligence" are mentioned, the translation will always be "AI". 

When translating from English to Spanish a video about product testing, the word "test" might get translated to "prueba", which is more of a classroom related test. So here we can force the translation for test to be "demostración" which is more accurate to our product testing.

Users can set the desired tone for translations, such as happy or upbeat, to ensure that the translated content aligns with their brand's voice. 

You can use Tones if you want to maintain the same style and accurately convey the tone and nuances of the original content. Words like:" Formal", "Technical", would be a good addiction here. 

This allows users to specify a vocabulary that should be recognized and pronounced correctly during translation. 

When trying to translate HeyGen brand name, it helps to add pronunciation here and write it as Hey-Jen to make sure it's pronounced correctly. 

The term "AI" would need to be pronounced "Aey-I" to accurately be translated in videos.

You can click on the play button to listen to the pronunciation and correct it. Click on the check mark to accept the word.

The Brand Voice feature is a powerful tool for users looking to maintain consistency and accuracy in their translations. By allowing users to specify untranslated words, enforce specific translations, and set the desired tone, Brand Voice ensures that translated content aligns with the brand's identity and message.

For further assistance, please contact our support team at 

This article helps explain the functions of the Brand Knowledge-based part of Video Translation

How to use Brand Knowledge-Based in Video Translation V1

How to use Brand Voice-Based in Video Translation

Why HeyGen Video translation feature is awesome?

Video Translate is an innovative tool designed to enhance communication by translating the speech in videos in real-time. The following diagram shows the old workflow of translating a video compared to the new one using HeyGen:

For optimal performance and to ensure the highest quality of lip sync translation, it is important to follow the guidelines provided below.

Video Translate performs best when there are up to two speakers visible on screen.

To ensure accurate voice capture and translation, speakers should face the camera directly, with their orientation no greater than 45 degrees away from facing the camera straight on. This positioning helps the product accurately capture audio and lip sync for effective video quality.

 For best results, speakers should be within 10 feet of the camera. This distance allows the product to effectively capture audio clarity and facial expressions, enhancing the translation accuracy and lip sync quality.

 Close-up shots of the speakers are preferred. Close-ups help in capturing detailed visual cues, which are essential for high quality lip sync.

Dynamic Shot Cuts: Video Translate does not perform well in scenarios with frequent and dynamic camera shot cuts. Such conditions can disrupt the continuous capture of audio and visual cues necessary for lip sync. To ensure optimal performance, maintain a steady shot focusing on the speakers.

Multiple Speakers Talking Over Each Other:

 The product's accuracy diminishes when multiple speakers talk simultaneously. For the most effective use of Video Translate, ensure that only one speaker talks at a time, allowing the product to accurately lip sync.

Background Noise: Minimize background noise to ensure the audio captured is as clear as possible. Excessive noise can interfere with speech recognition and lip sync accuracy.

By adhering to these guidelines, users can maximize the effectiveness of Video Translate, ensuring clear, accurate, and timely translation lip sync for videos. These practices are designed to enhance the lip sync quality by optimizing the conditions under which Video Translate operates.

The Expert - How to use Video Translation Feature

For users who have high bar for the translation accuracy, 

 can access additional features within the Video Translate tool.

: You can now preview the script after translation and make modifications in advance. Just upload your video and click the 'Generate Proofread' button allowing you to proof read the script. Our service will automatically analyze and translate the dialogue text and structure in your video, while also managing speaker recognition, even for videos with multiple speakers. You can then select the preferred voice for each speaker, enabling you to fine-tune the audio details and enhance the viewing experience of the translated video.

 - To enhance the accuracy of your translated content, you can upload the original subtitle files before starting the proofreading process. Additionally, you have the option to edit the translated scripts on the Proofread page. This allows you to ensure the quality and accuracy of the translated script before completion.

This article talks about how Enterprise users can utilize features like Proofread & Upload SRT to achieve higher quality translation results

What are Proofread & Upload SRT

When using HeyGen's Video Translation tool, your source video must meet certain requirements related to file size, type, & input language. These requirements can be found under the "Requirements" info button within the Video Translate tool.

We're working hard to add more supported languages based on our users feedback. Currently supported languages; 

Arabic, Catalan, Danish, Dutch, English, Filipino, Finnish, French, German, Hebrew, Hindi, Italian, Japanese, Korean, Mandarin, Norwegian, Polish, Portuguese, Romanian, Russian, Spanish, Swedish, Thai, Turkish, Ukrainian, Urdu, Vietnamese

Arabic, Bulgarian, Croatian, Czech, Danish, Dutch, English - American Accent, English - Your Accent, Filipino, Finnish, French, German, Greek, Hindi, Indonesian, Italian, Japanese, Korean, Malay, Mandarin, Polish, Portuguese, Romanian, Russian, Slovak, Spanish, Swedish, Tamil, Turkish, Ukrainian

Video files uploaded locally from your computer or linked from YouTube/Google Drive must be

File length:  Between 1 second & 1 hour in length

Google Drive URL: file must be shared w/ anyone who has the link