Five Companies Leading the Way in Deepfake Technology
Deepfake technology is a machine learning model that can interchange one person’s face with another in any video production. It was first introduced in 2018, and with advances in technology, it has been refined so you can’t always distinguish original videos from deepfakes. The model has become a means of generating fairly convincing forgeries. It has also caught on as a new type of machine learning that can be used for a variety of media applications including faking avatars, photographs, voices, and more. It’s become a legitimate option with many future useful possibilities in media. New startups are pursuing an interest in these options. Here are five companies that are leading the way in Deepfake technology for your consideration.
1. Topaz labs
Topaz Labs is a software startup that was found by Eric and Albert Yang, father and son owners. The company develops tools for editing images. Other products include plug-ins that can be used for editing in Adobe After Effects and other similar software. The software can pull images from video with harp images along with adding a range of social effects for moving images. The technology is built around machine learning featuring the use of an artificial intelligence-powered software package that allows conversion of JPEGs to RAW images. Their top-selling product is Gigapixel AI which has achieved the goal of enlarging low-resolution images to up to 600 percent larger with crystal clear sharpness. The latter is also useful for enhancement of scenery and landscape versus adding detail in photos, but Topaz Labs has it covered with other editing tools.
2. Modulate
Modulate is a deepfake startup that was established in 2018. The company launched with funding of $2 million in seed funding provided by Harmonix and others in the industry. Carter Huffman is the founder who partnered with a few friends from MIT to develop a program that offers deep fake technology for voices and audio. It allows you to substitute one voice for another. The platform developed by Modulate offers 3 unique components in one package. Voicewear allows you to hear voice skins and allows gamers to become their players by substituting their voices for the original characters in the game. The Toxmod component detects disruptive speech and stops it dead in its tracks. The third component is called VoiceVibe. It enhances engagement with community insights through the use of voice analytics technology. Module AI provides all three technologies in one suite.
There are a few potential downsides to the technology because it could be used to create problems with faked voices. Hopefully, reverse technology would be developed to detect the use of deepfake technology to resolve and mitigate these potential problems. This is one area of controversy that is likely to crop up in the future. This is sophisticated software that uses machine learning technology to substitute anyone’s voice for yours or vice versa.
3. Respeecher
Respeecher is a company that specializes in Voice Cloning technology that features emotional nuance and high quality. It’s similar to Modulate, yet targets a different audience. Respeecher is geared more towards the animation and film industries instead of the gaming community. Respeecher’s voice cloning technology has the potential to replace the voices of voice actors easily by recording a small sample of any voice to clone the tones, inflections, and other elements that make that voice unique. This is a product that could save time, effort, and money. Voice actors would no longer need to memorize and rehearse their lines. So far, the voice clones that Respeecher has generated have been impressive enough to draw the attention of a variety of investors, raising $1.5 million in startup funding.
4. Rephrase.ai
Rephrase.ai was founded in 2018 by Ashray Malhotra, Shivam Mangla, and Nisheeth Lahoti in Bangalore, Karnataka, India. The company was founded by non-equity assistance. Rephrase provides software for the creation of presentation videos that are entirely driven by artificial intelligence technology. The visual dubbing tool is also used for animations. The company’s deep learning engine can generate faces that are photorealistic to accompany speech, text, or audio by analyzing facial expressions and movements and generating them under the unique patterns that it recognizes. Information bout public presenters may be selected from a panel built into the software, or users may create their own presenters. What is so unique about this software is that the user has total control over the customization of their presenters. This text to video text allows for complete personalization of marketing and sales materials. It’s useful in a variety of industries. It is presently being utilized by businesses in automotive, financial institutions, and real estate. It is expected to become more popular in other areas of business.
5. D-ID
D-ID is a company that specializes in the development of solutions that protect identities from facial recognition. The company was founded in 2017 by Eliran Kuta, Sella Blondheim, and Gil Perry with its headquarters in Tel Aviv, Israel. D-ID’s products rely upon artificial intelligence and facial recognition technology to ensure the privacy of users. The company raised $22 million in Series A funding. Just as facial recognition software is designed to identify specific persons in a group, D-ID’s solution is software that de-identifies faces. It uses deep learning and artificial intelligence technologies to remove identifiable facial features without taking away key attributes including emotion, gender, and age. This software is compatible for use with videos and images. It’s a type of security that is intended to enhance security for biometric databases. The Smart Anonymizatioon component provides protection of sensitive personal data and it is also useful for protecting video data for analytics. The outlook is positive for this software to enhance security and protect databases from intrusion with the intent to perform the crime of identity theft. It’s intended for use by corporations in need of enhanced privacy and security.
