BodyTalk is Panjaya’s AI Dubbing Platform, which enables anyone to localize their video content for any audience, in any language, while preserving on-screen speakers’ natural voice and lip movements. BodyTalk produces localized videos that look and feel as natural as the original, through:

Automated Adaptive Translation, Speech Modulation, and Audio Retiming: BodyTalk automatically adjusts translations and aligns the rhythm and tone of the dubbed audio with the pauses, breaths, gestures and other body movements in the original video. This adaptive approach to translation and speech synthesis ensures that viewers experience the message as naturally and authentically as the original audience.

Precision Tools for Human-in-the-Loop Adjustment: We know that great localization requires nuance. BodyTalk features a suite of precision tools that give video creators the ability to modify translations or adjust performance nuances. Human-in-the-loop adjustment means you remain in control of your content's message, tone, and emotion.

Our automated solution makes it easy to dub in 3 simple steps. First, pick your target language(s) and upload your video. BodyTalk then creates a draft audio-only dub for each language. Second, review and, optionally, edit the translation or audio of the draft dubs in our Studio. Third, generate the final dubbed video with or without lip syncing and download.

BodyTalk supports dubbing in the following languages: Arabic, Bulgarian, Chinese, Czech, Danish, Dutch, English (US), Finnish, French, German, Greek, Hebrew (source only), Hindi, Italian, Japanese, Korean, Polish, Portuguese (BR), Portuguese (PT), Romanian, Russian, Slovak, Spanish, Spanish (MX), Swedish, Tamil, Turkish, Ukrainian.

Dubbing times vary and depend on the video length and complexity. A general rule of thumb - it takes about the length of the video to generate a dub ready for review, and about twice the length to incorporate lip syncing.

Though our automated platform accounts for all the complexities of dubbing, it can make mistakes. From time to time, there may be transcription or translation errors. Additionally, overlaps or general timing mismatches of the audio may occur due to differences in languages. Fortunately, our Studio provides a suite of tools for addressing these issues. Transcription and translation text can easily be edited in the Studio, and dub timing overlaps and mismatches can also be addressed through our audio editing tools.

Our customers own all intellectual property rights to all their uploaded and generated content. We will not use customer content independently to generate content of our own. We will only use customer content for marketing purposes with customers’ permission. However, we retain a license to use the uploaded and generated content to the degree necessary to run our services, as well as improve our models and services.

We are committed to the ethical use of our product. Our mission is to empower storytellers to easily adapt their video content to speak to any audience, anywhere.

However, we are acutely aware of the potential for misuse and are unequivocal in our belief that our platform should never be used to impersonate or harm others, nor play any part in spreading misinformation. To that end, we are developing both policy and technological measures to proactively combat misuse. Out terms strictly prohibit the use our platform to impersonate or harm others, including the generation of misinformation. We’re also investigating a number of technological tools, including those for detection and blocking of harmful content, as well as tools for identifying content created through our platform.