University of Washington researchers have developed new algorithms that solve a thorny challenge in the field of computer vision: turning audio clips into a realistic, lip-synced video of the person speaking those words.
As detailed in a paper to be presented Aug. 2 at SIGGRAPH 2017, the team successfully generated highly-realistic video of former president Barack Obama talking about terrorism, fatherhood, job creation and other topics using audio clips of those speeches and existing weekly video addresses that were originally on a different topic.
“These type of results have never been shown before,” said Ira Kemelmacher-Shlizerman, an assistant professor at the UW’s Paul G. Allen School of Computer Science & Engineering. “Realistic audio-to-video conversion has practical applications like improving video conferencing for meetings, as well as futuristic ones such as being able to hold a conversation with a historical figure in virtual reality by creating visuals just from audio. This is the kind of breakthrough that will help enable those next steps.”
In a visual form of lip-syncing, the system converts audio files of an individual’s speech into realistic mouth shapes, which are then grafted onto and blended with the head of that person from another existing video.
The team chose Obama because the machine learning technique needs available video of the person to learn from, and there were hours of presidential videos in the public domain. “In the future video, chat tools like Skype or Messenger will enable anyone to collect videos that could be used to train computer models,” Kemelmacher-Shlizerman said.
Because streaming audio over the internet takes up far less bandwidth than video, the new system has the potential to end video chats that are constantly timing out from poor connections.
“When you watch Skype or Google Hangouts, often the connection is stuttery and low-resolution and really unpleasant, but often the audio is pretty good,” said co-author and Allen School professor Steve Seitz. “So if you could use the audio to produce much higher-quality video, that would be terrific.”
By reversing the process — feeding video into the network instead of just audio — the team could also potentially develop algorithms that could detect whether a video is real or manufactured.
The new machine learning tool makes significant progress in overcoming what’s known as the “uncanny valley” problem, which has dogged efforts to create realistic video from audio. When synthesized human likenesses appear to be almost real — but still manage to somehow miss the mark — people find them creepy or off-putting.
“People are particularly sensitive to any areas of your mouth that don’t look realistic,” said lead author Supasorn Suwajanakorn, a recent doctoral graduate in the Allen School. “If you don’t render teeth right or the chin moves at the wrong time, people can spot it right away and it’s going to look fake. So you have to render the mouth region perfectly to get beyond the uncanny valley.”
Previously, audio-to-video conversion processes have involved filming multiple people in a studio saying the same sentences over and over to try to capture how a particular sound correlates to different mouth shapes, which is expensive, tedious and time-consuming. By contrast, Suwajanakorn developed algorithms that can learn from videos that exist “in the wild” on the internet or elsewhere.
“There are millions of hours of video that already exist from interviews, video chats, movies, television programs and other sources. And these deep learning algorithms are very data hungry, so it’s a good match to do it this way,” Suwajanakorn said.
Rather than synthesizing the final video directly from audio, the team tackled the problem in two steps. The first involved training a neural network to watch videos of an individual and translate different audio sounds into basic mouth shapes.
By combining previous research from the UW Graphics and Image Laboratory team with a new mouth synthesis technique, they were then able to realistically superimpose and blend those mouth shapes and textures on an existing reference video of that person. Another key insight was to allow a small time shift to enable the neural network to anticipate what the speaker is going to say next.
The new lip-syncing process enabled the researchers to create realistic videos of Obama speaking in the White House, using words he spoke on a television talk show or during an interview decades ago.
Currently, the neural network is designed to learn on one individual at a time, meaning that Obama’s voice — speaking words he actually uttered — is the only information used to “drive” the synthesized video. Future steps, however, include helping the algorithms generalize across situations to recognize a person’s voice and speech patterns with less data – with only an hour of video to learn from, for instance, instead of 14 hours.
“You can’t just take anyone’s voice and turn it into an Obama video,” Seitz said. “We very consciously decided against going down the path of putting other people’s words into someone’s mouth. We’re simply taking real words that someone spoke and turning them into realistic video of that individual.”
The Latest on: Audio-to-video conversion
- How Motherhood Inspired This CEO to Use Computer Vision to Stop Mass Shootingson October 29, 2019 at 8:56 am
So Falzone combined her tech-entrepreneurial experience with her desire to prevent crime, co-founding Athena Security, the first camera surveillance system to use computer vision to spot guns ...
- Facebook AI Releases New Computer Vision Library Detectron2on October 28, 2019 at 3:21 am
Facebook AI Research (FAIR) has released Detectron2, a PyTorch-based computer vision library that brings a series of new research and production capabilities to the popular framework. Since its ...
- An Inside Look at Voice-enabled Edge Computing and Computer Visionon October 25, 2019 at 10:18 am
We call this Smart Edge AI and it involves our powerful SoCs with neural network accelerators combined with intelligent computer vision and voice processing. I think one of the greatest aspects about ...
- Overview of Computer Vision (CV) and Machine Vision (MV) Technology Applications in Everyday Life, 2019 Study - ResearchAndMarkets.comon October 22, 2019 at 5:28 pm
The "Computer Vision and Machine Vision in Everyday Life" report has been added to ResearchAndMarkets.com's offering. The Report Includes: An overview of application specific markets of computer ...
- Disperse raises $15 million for computer vision that tracks construction site progresson October 22, 2019 at 12:13 am
At its core, Disperse is all about helping companies track progress across construction projects to identify inefficiencies, through capturing imagery and using computer vision to spot issues as soon ...
- Computer Vision Can Transform Educationon October 18, 2019 at 11:32 pm
The education sector has long treated every student the same. However, every student is unique and has different learning capabilities. The use of computer vision in education can help to maximize ...
- Computer Vision and Machine Vision in Everyday Lifeon October 18, 2019 at 11:14 am
New York, Oct. 18, 2019 (GLOBE NEWSWIRE) -- Reportlinker.com announces the release of the report "Computer Vision and Machine Vision in Everyday Life" - ...
- Presto Launches Computer Vision Product to Deliver Ground-Breaking, Real-Time Actionable Insights to Restaurantson October 16, 2019 at 5:00 am
REDWOOD CITY, Calif.--(Business Wire)--Presto, provider of the restaurant industry’s next generation front-of-house (FOH) technology platform, announces the launch of Presto Vision -- the first ...
- Global AI In Computer Vision Market 2019-2025 Top Key Players- Qualcomm, Google, Basler and moreon October 16, 2019 at 2:36 am
Oct 16, 2019 (Heraldkeepers) -- AI In Computer Vision MarketThis report focuses on the global AI In Computer Vision status, future forecast, growth opportunity, key market and key players. The study ...
- Powerful computer vision algorithms are now small enough to run on your phoneon October 11, 2019 at 10:04 am
Researchers have shrunk state-of-the-art computer vision models to run on low-power devices. Growing pains: Visual recognition is deep learning’s strongest skill. Computer vision algorithms are ...
via Google News and Bing News