| |
|
|
| |
|
|
Scientists Train AI To Learn People's Voices, Then Generate Their Faces
An neural network named "Speech2Face" was trained by scientists on millions of educational videos from the internet that showed over 100,000 different people talking. From this dataset, Speech2Face learned associations between vocal cues and certain physical features in a human face, researchers wrote in a new study. The AI then used an audio clip to model a photorealistic face matching the voice, and the results are surprisingly close to the actual faces of the people whose voices it listened to. The faces generated by Speech2Face didn't precisely match the people behind the voices. But the images did usually capture the correct age ranges, ethnicities and genders of the individuals, according to the study.
The findings have been published in the preprint journal arXiv but have not been peer-reviewed. |
|
|
|
| |
|
| There are no conversations. |
| |
|
|
| cauz |
June 12, 2019, 8:29 a.m. |
| |
|
|
|
| Chris Hadfield |
Some of the greatest reality television we ever had was the moon landings. When you think about it, that was human emotion and people, unscripted, working with each other - and millions and millions of people around the world, glued to their television sets to share real-time in a brand new, fascinating human experience. |
| Vanilla Ice |
One thing I didn't understand in life was that I had $100,000,000 in the bank and I couldn't buy happiness. I had everything: mansions, yachts, Ferraris, Lamborghinis, but I was depressed. I didn't know where I fitted in. But then I found family and friends and I learned the value of life. |
| Dyan Cannon |
They offered me millions and millions and millions of dollars to write books about Cary. That was between us. That was private. I'll always love him. |
| Sebastian Bach |
My vocal style is called bel canto, which is an old Italian vocal style going back hundreds of years. |
| Sheila E. |
If you are on television, that one night that someone is able to see you, you're talking about millions of people in one day. That's pretty awesome because then it helps you to jumpstart your career, and if you're the one to go out there and perform, people are going to come see you because those millions saw you on the television. |
| Spencer Bachus |
Shockingly, a University of Pennsylvania study says the number of young people addicted to gambling - largely due to increased exposure to the Internet and Internet gambling - grew by an alarming 20 percent between 2004 and 2005 alone. |
| Gilbert K. Chesterton |
The object of a New Year is not that we should have a new year. It is that we should have a new soul and a new nose; new feet, a new backbone, new ears, and new eyes. Unless a particular man made New Year resolutions, he would make no resolutions. Unless a man starts afresh about things, he will certainly do nothing effective. |
| Spencer Bachus |
In fact, a University of Connecticut study showed that as many as three in four pre-teens and teens who are exposed to Internet gambling become addicted. |
| David Icke |
The families in positions of great financial power obsessively interbreed with each other. But I'm not talking about one Earth race, Jewish or non-Jewish. I'm talking about a genetic network that operates through all races, this bloodline being a fusion of human and reptilian genes. |
| Matt Dallas |
ABC Family is really restructuring their network because there's a new kind of family, so I think they're really trying to step away from that younger audience, and they're bringing a network that can bring more to an older demographic. |
|
|
|
|
Researchers Have Figured Out How To Fake News Video With AI
A team of computer scientists at the University of Washington have used artificial intelligence to render visually convincing videos of Barack Obama saying things he's said before, but in a totally new context. In a paper published this month, the researchers explained their methodology: Using a neural network trained on 17 hours of footage of the former U.S. president's weekly addresses, they were able to generate mouth shapes from arbitrary audio clips of Obama's voice. The shapes were then textured to photorealistic quality and overlaid onto Obama's face in a different "target" video. Finally, the researchers retimed the target video to move Obama's body naturally to the rhythm of the new audio track. In their paper, the res...
|
| |
|
|
|
Google Works Out a Fascinating, Slightly Scary Way For AI To Isolate Voices In a Crowd
https://arstechnica.com/gadgets/2018/04/google-works-out-a-fascinating-slightly-scary-way-for-ai-to-isolate-voices-in-a-crowd/
An anonymous reader quotes a report from Ars Technica:
Google researchers have developed a deep-learning system designed to help computers better identify and isolate individual voices within a noisy environment. As noted in a post on the company's Google Research Blog this week, a...
|
| |
|
|
|
Mozilla Releases Open Source Speech Recognition Model, Massive Voice Dataset
Mozilla's VP of Technology Strategy, Sean White, writes:
I'm excited to announce the initial release of Mozilla's open source speech recognition model that has an accuracy approaching what humans can perceive when listening to the same recordings... There are only a few commercial quality speech recognition services available, dominated by a small number of large companies. This reduces user choice and available features for startups, researchers or even larger companies that want to speech-enable their products and services. This is why we started DeepSpeech as an open source project.
...
|
| |
|
|
|
AI System Detects ‘Deception’ in Courtroom Videos
Analyzing facial micro-expressions yields almost 90 percent accuracy in detecting lies.
Courtrooms are inexact places. Juries and the processes they use to reach verdicts are parameterized, but a trial is nonetheless all about convincing those juries of something that is inexact and subjective at its core. In the US, this is “reasonable doubt.” To find guilt, a judge and-or jury must determine beyond a reasonable doubt that a defendant is guilty of an offense. In one courtroom, that reasonable doubt maybe be countered by overwhelming physical evidence, while, in another, it may be the testimony of an incentivized witness.
...
|
| |
|
|
|
Bees Can Link Symbols To Numbers, Study Finds (phys.org)
Researchers have trained honeybees to match a character to a specific quantity, revealing they are able to learn that a symbol represents a numerical amount. The discovery, from the same Australian-French team that found bees get the concept of zero and can do simple arithmetic, also points to new approaches for bio-inspired computing that can replicate the brain's highly efficient approach to processing. Associate Professor Adrian Dyer said while humans were the only species to have developed systems to represent numbers, like the Arabic numerals we use each day, the research shows the concept can be grasped by brains far smaller ...
|
| |
|
|
|
Lyrebird claims it can recreate any voice using just one minute of sample audio
Artificial intelligence is making human speech as malleable and replicable as pixels. Today, a Canadian AI startup named Lyrebird unveiled its first product: a set of algorithms the company claims can clone anyone’s voice by listening to just a single minute of sample audio.
A few years ago this would have been impossible, but the analytic prowess of machine learning has proven to be a perfect fit for the idiosyncrasies of human speech. Using artificial intelligence, companies like Google have been able to creat...
|
| |
|
|
|
This clever AI hid data from its creators to cheat at its appointed task
https://techcrunch.com/2018/12/31/this-clever-ai-hid-data-from-its-creators-to-cheat-at-its-appointed-task/
Depending on how paranoid you are, this research from Stanford and Google will be either terrifying or fascinating. A machine learning agent intended to transform aerial images into street maps and back was found to be cheating by hiding information it would need later in “a nearly imperceptible, high-frequency signal.” Clever girl!
...
|
| |
|
|
|
|
" among 20 British children aged five, researchers concluded that among the children in the study, the children who avoid eye contact while considering their responses to questions are more likely to answer correctly than children who maintain eye contact. While humans obtain useful information from looking at the face when listening to someone, the process of looking at faces is mentally demanding and takes processing. Therefore, it may be unhelpful to look at faces when trying to concentrate and process something else that's mentally demanding." Finally something reaffirming to the fact I cant listen to someone if I'm trying to maintain eye contact.
|
| |
|
|
|
A 2013 study from Emory University found that mice trained to fear a specific odor would pass their emotions on to their offspring and future generations. Scientists applied electric shocks to mice as they exposed them to the smell of cherry blossoms. The mice then bred, and both the children and grandchildren of the affected rodents demonstrated a fear of cherry blossoms the first time they smelled them.
“Our results allow us to appreciate how the experiences of a parent, before even conceiving offspring, markedly influence both structure and function in the nervous system of subsequent generations,” Dr. Brian Dias of the Emory University department of psychiatry said to the Daily Telegraph. “Such a phenomenon may contribute to the etiology and potential intergenerational transmission of risk for neuropsychiatric disorders such as phobias, anxiety and post-traumatic stress disorder.”
...
|
| |
|
|
|
Prisons Across the US Are Quietly Building Databases of Incarcerated People's Voice Prints (theintercept.com)
In New York and other states across the country, authorities are acquiring technology to extract and digitize the voices of incarcerated people into unique biometric signatures, known as voice prints. From a report:
Prison authorities have quietly enrolled hundreds of thousands of incarcerated people's voice prints into large-scale biometric databases. Computer algorithms then draw on these databases to identify the voices taking part in a call and to search for other calls in which the voices of interest are detected. Some programs, like New York's, even analyze the voices of call...
|
| |