Synthetic Media & Generative AI News and Discussions

User avatar
Yuli Ban
Posts: 4631
Joined: Sun May 16, 2021 4:44 pm

Re: Synthetic Media & Deepfakes News and Discussions

Post by Yuli Ban »

And remember my friend, future events such as these will affect you in the future
User avatar
caltrek
Posts: 6474
Joined: Mon May 17, 2021 1:17 pm

Re: Synthetic Media & Deepfakes News and Discussions

Post by caltrek »

Here's Why We're About to See an Explosion of Hyperreal Artificial Humans Online
by Mike Butcher
April 25, 2022

https://techcrunch.com/2022/04/25/heres ... ns-online/

Introduction:
(TechCrunch) Synthetically generated versions of real people that can be can be programmed to say anything sounds like a scenario from the latest episode of “Black Mirror.”

But in fact, production-grade video-based characters based on real people — which can talk about any product or subject at all, in a hyperlifelike manner — are arguably going to be part of the next wave in areas like e-commerce and remote learning. Further, a Hollywood celebrity could simply license out their avatar to explain products, at a scale that would make it impossible to physically film. But perhaps more realistically, “digital twins” like this make much more convincing videos than invented characters, because of their humanlike qualities.

The market for this technology is expanding. Key players in the space include SoulMachines (which has raised $135 million) and Synthesia (raised $66.6 million).

Back in 2020 we reported how Hour One, a New York and Tel Aviv startup which creates AI-driven synthetic characters based on real humans, had closed a $5 million seed funding round.

It’s now raised a $20 million Series A funding round led by Insight Partners. Also participating in the round was Galaxy Interactive, Remagine Ventures, Kindred Ventures, Semble Ventures, Cerca Partners, Digital-Horizon and Eynat Guez.
Don't mourn, organize.

-Joe Hill
User avatar
caltrek
Posts: 6474
Joined: Mon May 17, 2021 1:17 pm

Re: Synthetic Media & Deepfakes News and Discussions

Post by caltrek »

Video of Stricken Russian Warship on Fire: Fake or Real?
by Thomas Newdick
May 6, 2022

https://www.thedrive.com/the-war-zone/u ... ip-on-fire

Extract:
(The Drive) Ukrainian officials claim that the Russian Navy’s Admiral Grigorovich class frigate Admiral Makarov, one of the most advanced in its fleet, has become the latest victim of Ukrainian anti-ship missiles while operating in the Black Sea.

While unverified imagery has begun to appear on social media, purporting to show the frigate ablaze, there are some serious questions about its authenticity:

The warship seen in the above footage does appear to look broadly like an Admiral Grigorovich class, or at least another member of the Krivak class from which it was derived. Nonetheless, it has been met with skepticism, with suggestions that it may be a doctored or composite video. This would not be the first time that falsified video has been used in support of Ukrainian claims in the conflict.
Don't mourn, organize.

-Joe Hill
User avatar
Yuli Ban
Posts: 4631
Joined: Sun May 16, 2021 4:44 pm

Re: Synthetic Media & Deepfakes News and Discussions

Post by Yuli Ban »

NaturalSpeech: End-to-End Text to Speech Synthesis with Human-Level Quality
Text to speech (TTS) has made rapid progress in both academia and industry in recent years. Some questions naturally arise that whether a TTS system can achieve human-level quality, how to define/judge human-level quality and how to achieve it. In this paper, we answer these questions by first defining the criterion of human-level quality based on statistical significance of measurement and describing the guidelines to judge it, and then proposing a TTS system called NaturalSpeech that achieves human-level quality on a benchmark dataset. Specifically, we leverage a variational autoencoder (VAE) for end-to-end text to waveform generation, with several key designs to enhance the capacity of prior from text and reduce the complexity of posterior from speech, including phoneme pre-training, differentiable duration modeling, bidirectional prior/posterior modeling, and memory mechanism in VAE. Experiment evaluations on popular LJSpeech dataset show that our proposed NaturalSpeech achieves

0.01
CMOS (comparison mean opinion score) to human recordings on sentence level, with Wilcoxon signed rank test at p-level
p

0.05
, which demonstrates no statistically significant difference from human recordings for the first time on this dataset.
Try out the audio samples!
And remember my friend, future events such as these will affect you in the future
User avatar
Yuli Ban
Posts: 4631
Joined: Sun May 16, 2021 4:44 pm

Re: Synthetic Media & Deepfakes News and Discussions

Post by Yuli Ban »

And remember my friend, future events such as these will affect you in the future
User avatar
Yuli Ban
Posts: 4631
Joined: Sun May 16, 2021 4:44 pm

Re: Synthetic Media & Deepfakes News and Discussions

Post by Yuli Ban »

What DALL-E 2 can and cannot do
I got access to DALL-E 2 earlier this week, and have spent the last few days (probably adding up to dozens of hours) playing with it, with the goal of mapping out its performance in various areas – and, of course, ending up with some epic art.

Below, I've compiled a list of observations made about DALL-E, along with examples. If you want to request art of a particular scene, or to test see what a particular prompt does, feel free to comment with your requests.
DALL-E's strengths
• Stock photography content
• Pop culture and media
• Art style transfer
• Creative digital art
• The future of commercials
DALLE's weaknesses
• Scenes with two characters
• Foreground and background
• Novel objects, or nonstandard usages
• Spelling
• Realistic human faces
• Limitations of the "edit" functionality
And remember my friend, future events such as these will affect you in the future
User avatar
Yuli Ban
Posts: 4631
Joined: Sun May 16, 2021 4:44 pm

Re: Synthetic Media & Deepfakes News and Discussions

Post by Yuli Ban »

And remember my friend, future events such as these will affect you in the future
User avatar
Yuli Ban
Posts: 4631
Joined: Sun May 16, 2021 4:44 pm

Re: Synthetic Media & Deepfakes News and Discussions

Post by Yuli Ban »

Can AI write an episode of Stargate? Google AI took on the challenge
The process of writing a television show typically involves a writers room and a lot of time, as humans figure out the plot and the dialogue that makes a show work.

For the cult classic Stargate science fiction franchise, which spanned three series (SG-1, Stargate Atlantis and Stargate Universe), character and plot development was helmed by Stargate co-creator Brad Wright. In 2021, Wright publicly posted a message on Twitter asking if it was possible for AI to write an episode of Stargate that would appear on SciFi insider site The Companion.

None other than Laurence Moroney, AI lead at Google, responded by picking up the gauntlet to try and prove what AI could do.– though he wasn’t initially worried that AI would replace him or other writers.

“When the whole project started, I threw it out there to The Companion as an idea – I knew I had seen a couple of AI models for scripts,” Wright told VentureBeat. “In some ways, it’s fantastic. In other ways, it’s very unthreatening.”

The first iteration of the AI-generated script was completed by November 2021. The script was interesting, but there was also a lot of gibberish, Wright said. Now Moroney and Wright are collaborating again on a second version that aims to dial a new gate address for a more involved and engaging Stargate script.
And remember my friend, future events such as these will affect you in the future
User avatar
Yuli Ban
Posts: 4631
Joined: Sun May 16, 2021 4:44 pm

Re: Synthetic Media & Deepfakes News and Discussions

Post by Yuli Ban »

And remember my friend, future events such as these will affect you in the future
User avatar
Yuli Ban
Posts: 4631
Joined: Sun May 16, 2021 4:44 pm

Re: Synthetic Media & Deepfakes News and Discussions

Post by Yuli Ban »



And remember my friend, future events such as these will affect you in the future
Post Reply