Videos

Synthetic voices have become ubiquitous. They feed us directions in the morning, shepherd us through phone calls by day and broadcast the news on smart speakers at night. And as the technology used to make them improves, these voices are becoming more and more human-sounding. This is the final frontier in synthetic speech: replicating not just what we say but how we say it.

Rupal Patel heads a research group at Northeastern University that studies speech prosody—the changes in pitch, loudness and duration that we use to convey intent and emotion through voice. “Sometimes people think of it as the icing on the cake,” she explains. “You have the message, and now it’s how you modulate that message, but I really think it’s the scaffolding that gives meaning to the message itself.”

Patel says she grew interested in prosody after finding it was the only element of vocal communication that seemed to be available to people with some kinds of severe speech disorders. These patients were able to make expressive sounds even if they could not speak clearly. In 2014 Patel founded a company to build custom synthetic voices for nonspeaking individuals. VocaliD has since expanded to commercial brands and influencers.

Synthetic speech has come a long way over the years. At age nine, Siri is the oldest virtual assistant—but in the world of speaking machines, she’s a baby. People have been trying to synthesize speech since at least the 18th century, when an Austro-Hungarian inventor built a crude replica of the human vocal tract that could articulate entire phrases (albeit in a monotone).

Current machine-learning techniques can model human speech, complete with awkward pauses and lip smacks. Still, training on thousands of samples per second is prohibitively expensive for most real-world systems. Researchers, including those at VocaliD, are continually implementing newer and more efficient methods.

But even as the remaining gaps between human and synthetic speech are steadily closing, truly lifelike prosody continues to elude even the most sophisticated systems. Maybe what’s still missing requires machines not only to mimic humans but also to feel like us.

Join Scientific American for a conversation about the next steps in humanity’s reconnaissance of Mars. Featuring Casey Dreier, senior space policy adviser at the Planetary Society, and space and physics editor Lee Billings, this deep dive will begin with an overview of NASA’s upcoming Perseverance rover—slated to land on Mars in February 2021 to search for signs of past and present life and to gather samples for future return to Earth.

Dreier and Billings will also discuss the “post-Perseverance” future in which space agencies and private companies may pursue major shifts in Mars exploration strategies and how those plans could forever change our understanding of—and relationship with—the Red Planet.

Sridhar Ravi was outdoors with his colleagues on a summer day in Germany when a group of bumblebees grabbed his attention.

As the bees made their way from flower to flower, they skillfully flew between obstacles, dodging branches and shrubs. These actions seemed to require a complex awareness of one's physical body in relation to one’s environment that had only been proven to exist in animals with large brains. 

To examine this, a team of researchers at Australia’s University of New South Wales, Canberra, led by Ravi, set up a hive of bumblebees inside their laboratory. The bees could come and go via a tunnel, which could be partially blocked with an adjustable barrier. Ravi and his team made the gap progressively smaller over time, and observed how the bees’ reactions changed.

The study, published in the Proceedings of the National Academy of Sciences, found the bumblebees measured the gap by flying side-to-side to scan it. When the gap became narrower than their wingspan, the bees took a longer time to scan the opening. And then they did something remarkable: they turned their bodies to fly through sideways. Some of the bees’ bodies did bump the sides of the narrowed opening—but every one of the 400 recorded flights through the gap was a success.

“Over thousands of years nature has coded insects with some amazing attributes,” Ravi says. “Our challenge now is to see how we can take this and apply similar coding to future robotic systems, enhancing their performance in the natural world.”

To understand the battle over e-cigarettes, one must look to history—and big tobacco

Earthly matters got you stressed? Here is an opportunity to elevate your mind above the terrestrial din with three cosmic questions

We probably think we know gravity pretty well. After all, we have more conscious experience with this fundamental force than with any of the others (electromagnetism and the weak and strong nuclear forces). But even though physicists have been studying gravity for hundreds of years, it remains a source of mystery.

In our video Why Is Gravity Different? We explore why this force is so perplexing and why it remains difficult to understand how Einstein’s general theory of relativity (which covers gravity) fits together with quantum mechanics.

Gravity is extraordinarily weak and nearly impossible to study directly at the quantum level. We cannot scrutinize it using particle accelerators like we can with the other forces, so we need other ways to get at quantum gravity.

Enter black holes. In a paper in the early 1970s the late physicist Jacob Bekenstein investigated the question of what happens to entropy—a measure of disorder, or randomness, in a system—when matter succumbs to a black hole’s massive gravitational pull and falls through its event horizon.

Bekenstein noted that the matter’s entropy seems to disappear inside the black hole.  Yet this would violate the second law of thermodynamics, which states two things: information cannot be destroyed, and entropy can only increase.  Thus the entropy of the black hole must compensate for the loss.  Bekenstein argued that this black hole entropy must not be proportional to the black hole’s volume, but to the area of its event horizon.

If we are describing the contents of black holes in terms of area instead of volume, we should think about laws of physics in terms of area as well. This would mean a theory of everything (gravity included) should be able to play out in fewer than three spacetime dimensions.

Now let’s imagine the information that describes the state of the entire universe—all stored on a single hard drive. And then throw that hard drive into a black hole. The stored information cannot be lost, so it must be contained in the surface area of the black hole (albeit scrambled).

This scenario leads to a dramatically new way of thinking in which the universe could effectively be a hologram, a seemingly 3-D object that is actually just a projection from a 2-D surface. Our ostensibly three-dimensional experience of the world would then be an illusion, convincingly generated by a fundamentally lower-dimensional reality.

Maybe we’re all just paper-thin cutouts drifting in gravity’s cosmic breeze.

What do zebras have to do with the structure of the cosmos? Imagine a single zebra in your mind. With twitching ears, tufted hair, and a visual interference pattern wrapped over muscle and skin, the animal has its own contours, which are easy to make out up close. But get a large group of zebras together (a dazzle, by definition), and the individuals blend and meld into a seemingly homogeneous sea of heads, hoofs and stripes, especially when viewed from a great distance. The matter in the universe, it seems, dazzles in much the same way.

In this video, How the Big Bang Governs the Texture of Our Universe, we explore the concept of cosmic inhomogeneities—using zebras as an analogy for clumps of matter scattered throughout the universe. Animator Lottie Kingslake depicts these whimsical and familiar characters as they graze and form larger and larger groups similarly to how matter is pulled together in galaxies. We were inspired by the color palette of Marylyn Dintenfass’s paintings, which make up the backdrop in the live-action portion of the video.

This visual rendering takes up a fundamental question about the texture of space: Is it smooth or bumpy? The answer is that it is both. It is homogeneous, or smooth, when viewed from vast distances but inhomogeneous, or bumpy, at the scale of planets and galaxies. This difference in texture gives us hints about our universe’s earliest history and even lets us characterize how dark matter is distributed.

But what exactly does it mean for the universe to be textured as it is? Empty space is smooth because there is no matter to alter the equilibrium of the universe. Moments after the big bang, there was a period of inflation, in which the universe expanded exponentially. At this point, the universe was almost perfectly smooth—because although it was devoid of stars and galaxies (the first atoms did not form until about 380,000 years after the big bang), there were tiny quantum ripples in spacetime.

Over time, these tiny quantum fluctuations were amplified by gravity. Clumps and voids formed, creating a bumpy texture on relatively small scales. As gravity pulled the clumps together, they collapsed into themselves to form stars, planets and galaxies—and inhomogeneity gradually increased over cosmological time.

Our universe remains smooth, or homogeneous, over very long distances, where gravitational collapse does not hold sway. But if you zoom way out, all of the matter and energy in the cosmos appears evenly distributed.

That’s a little mental roughage to think about the next time you find yourself at the zoo.

Life’s up and downs may seem as inevitable as gravity, but somehow 2020 feels worse than usual. Just as a thought experiment, what if this year actually did get so weird that it even ushered in a change in how gravity affects our material universe?

In the video Did the Universe Have to Be the Way That It Is? we examine what our universe—and more specifically, our lives—might look like with some tweaks to the physics responsible for the world as we know it.

If gravity were just a little stronger in our own three-dimensional world, the curvature of spacetime would be greater, and matter could more easily collapse in on itself. This arrangement would make stars, galaxies and planets extremely diminutive, compared with the ones in our reality. Not only would we have less space on Earth, but our sun would deplete its nuclear fuel much more rapidly—meaning that evolution, and life itself, would be greatly curtailed.

If gravity were weaker, Earth would be gigantic, and it might be oddly shaped like some asteroids—or a potato. And rather than walking on the surface of our planet, we might find ourselves, say, jumping to grab a rebound in a basketball game, only to accidentally end up in the upper atmosphere or orbiting the globe as a tiny human space station.

What if we weren’t three-dimensional at all? (Imagine people as paper cutouts.) If we lived in two dimensions, gravity would act very differently. Although we would still have the spacetime curvature noted in Einstein’s general theory of relativity, such curvature would no longer produce gravitational forces. For this sort of flattened universe, we could instead have “scalar gravity,” in which Newton’s description of gravity would have been the final word, and black holes would be relegated to science fiction.

Thankfully, even though 2020 seems topsy-turvy, we still have gravity to keep us grounded.

Four hundred years ago Galileo Galilei’s scientific findings were rejected because they didn’t fit the prevailing beliefs of the time. His story is disturbingly relevant today. Astrophysicist and author Mario Livio and Scientific American editor Clara Moskowitz to discusses lessons from Galileo’s life for dealing with science deniers now, plus a historical detective story about Galileo’s famous motto, “And yet it moves.”

A new system called PiVR creates working artificial environments for small animals such as zebra fish larvae and fruit flies. Developers say the system’s affordability could help expand research into animal behavior. 

Over the past few months, the phrase “social distancing” has entered our lexicon. Many of us have found ourselves separated from family and friends—or at least from our normal social lives. As humans grapple with pandemic-induced isolation, science is starting to offer insight into what may be happening in our brains when our social contact with others is dramatically reduced.

That insight happens to come from a place with more penguins than people.

Tim Heitland of the Alfred Wegener Institute Helmholtz Center for Polar and Marine Research in Germany spent 14 months in Antarctica between 2016 and 2018. When he returned, daily life felt overwhelming—everything from the colors and vegetation to all the other people. Part of the shock may have come from returning with a different brain than the one he left with.

While the members of Heitland’s crew conducted research on the earth’s iciest continent, they themselves were also being studied by researchers interested in how extreme work environments trigger neurological changes. The investigation showed that most of the people in Heitland’s team lost volume in parts of their hippocampus, a brain region involved in spatial navigation, learning and emotional processing. The phenomenon is similar to what scientists believe happens to prisoners in solitary confinement, where social isolation and sensory deprivation can lead to post-traumatic stress disorder.

This research appears particularly relevant now, when vast numbers of people are spending more and more time alone. Some scientists hope the work will lead to interventions that counteract the damage of isolation before it causes long-term problems.

Artificial intelligence still needs to bridge the “sim-to-real” gap. Deep-learning techniques that are all the rage in AI log superlative performances in mastering cerebral games, including chess and Go, both of which can be played on a computer. But translating simulations to the physical world remains a bigger challenge.

A robot named Curly that uses “deep reinforcement learning”—making improvements as it corrects its own errors—came out on top in three of four games against top-ranked human opponents from South Korean teams thatincluded a women’s team and a reserve squad for the national wheelchair team. (No brooms were used).

One crucial finding was that the AI system demonstrated its ability to adapt to changing ice conditions. “These results indicate that the gap between physics-based simulators and the real world can be narrowed,” the joint South Korean-German research team wrote in Science Robotics on September 23.