In This Update

• Support for Spanish and Portuguese audio
• Download videos from social platforms
• New and faster user interface
• Improved audio models with greater accuracy.

The Reality Defender Team is excited to announce the latest update to our award-winning deepfake detection models. This extensive new release builds upon our commitment to upgrade, iterate, and improve our numerous multimodal detection models, delivering cutting-edge detection that empowers clients and users to detect and mitigate the spread of deepfakes and manipulated media.

Detecting Multilingual Deepfakes

We made significant strides towards expanding our audio model deepfake capabilities, introducing support for Spanish and Portuguese audio files. Previously, our models performed best on English audio files, but with this new update, we have stellar performance across multiple languages. Deepfakes are a global issue, and this is our first step towards expanding our audio model coverage. 

Supporting Direct Social Media Downloads

Many of our customers monitor social media platforms for potential deepfakes that generate disinformation or brand risk. However, it’s often difficult to download the source file from these social media platforms. The Reality Defender web platform can now pull media files directly from popular social media sites, including YouTube, Twitter, TikTok, Instagram, and Facebook. All users have to do is provide a link and the Reality Defender web platform will download the associated video file at the highest possible resolution to detect potential manipulations.

Detect Deepfakes With a New User Interface

We completely refreshed the Reality Defender web app UI to provide a more intuitive, user-friendly, and perhaps most importantly, faster experience. Whether you're a seasoned user or just starting your journey with our platform, our new interface ensures that you can quickly and easily access the features you need to stay ahead of the evolving threat landscape.

Continued Improvements to Audio Models

In addition to supporting more languages, our audio models have been updated to detect the latest models. We’ve also adjusted our ensemble model to more aggressively flag suspicious files. Our ensemble model uniquely combines results from multiple models to make a final analysis of an audio file. 

What This Means for Users

The latest update to Reality Defender's deepfake detection platform represents a significant leap forward in our mission to safeguard the integrity of media. With enhanced audio capabilities, social media downloads, and a new UI, we're providing users with the most advanced tools available to detect and mitigate deepfakes — all from the most trusted name in detecting deepfakes.

To learn more about how Reality Defender can help you protect your organization from manipulated media, contact us today.