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 I built a fully functional Deep Learning Neural Network chatbot platform (NLU Engine) in under a week…. on July 28, 2017 at 4:54 am
For the last year or so I have been working on a number of Artificial Intelligence programs, covering computer vision and natural linguistics amongst other projects. In the field of natural linguistics I have built two main projects, a chatbot engine ... […]
- Artificial intelligence can turn 2D photos into real-world objects on July 27, 2017 at 10:39 pm
the authors reported this week at the Institute of Electrical and Electronics Engineers Conference on Computer Vision and Pattern Recognition, in Honolulu. The new program, called SurfNet (after the word “surface”), could also invent brand new ... […]
- Learn Artificial Intelligence with 10-Course Machine Learning Bundle on July 27, 2017 at 10:30 pm
You will learn many programming languages such as Java, Python, Hadoop, MapReduce for AI and data science. You will also be introduced to statistics and data science in R, deep learning and computer vision plus so much more. […]
- eBay’s New Computer Vision Program Will Let You Search for Products Using Images on July 27, 2017 at 8:09 am
Earlier today, online shopping platform eBay announced its plans to launch two new computer vision features: Find it on eBay and Image Search. In the same way Shazam created a useful new way to identify music just by hearing it, eBay has created these ... […]
- Classifying traffic signs with Apache MXNet: An introduction to computer vision with neural networks on July 27, 2017 at 6:03 am
The full Jupyter Notebook for this post can be accessed here. Although there are many deep learning frameworks, including TensorFlow, Keras, Torch, and Caffe, Apache MXNet in particular is gaining popularity due to its scalability across multiple GPUs. […]
- Agtech startup Prospera, which uses AI and computer vision to guide farmers, harvests $15M on July 26, 2017 at 9:29 pm
Tel Aviv-based startup Prospera has raised a $ 15 million Series B to expand the scope of its technology, which uses computer vision and artificial intelligence to help farmers analyze data gathered from their fields. The round was led by Qualcomm Ventures ... […]
- Las Vegas Looks to Expand Computer Vision Technology Pilot Program on July 26, 2017 at 2:49 pm
Las Vegas is in talks to expand a pilot program that uses sensors with computer vision technology to count vehicle and pedestrian traffic, thereby creating a sort of Google Analytics for information about the city’s streets. This program is made possible ... […]
- New Computer Vision SDK Beta R2 Brings Enhanced Deep Learning Capabilities, Supports More OSs, & Improves Performance on July 26, 2017 at 11:30 am
For computer vision, deep learning, video solutions (such as security cameras), or office automation, top systems may incorporate multi-channel streaming, real-time software-based analytics, and more. On the heels of its newest release, you can get your ... […]
- eBay announces computer vision search that helps you find items using photos on July 25, 2017 at 8:41 pm
With artificial intelligence (AI) and machine learning seemingly permeating just about every facet of the online world, it comes as little surprise to learn that eBay has some similar tricks up its proverbial sleeve. The ecommerce juggernaut has announced ... […]
- How NVIDIA AI labs are driving computer vision's future on July 25, 2017 at 1:05 pm
How is sports strategy like self-driving cars and brain tumor diagnosis? They're all the work of world-leading universities that are breaking new ground in artificial intelligence at the NVIDIA AI Labs. And they'll all be on deck in the next few days at ... […]
via Google News and Bing News