Google improved WaveNet, its text-to-speech AI, by feeding it Mel spectrograms instead of raw text. Trained on 24 hours of speech, the system got so good that listeners often couldn't tell the synthetic voice from a real one. The catch: it still can't generate audio in real time, and the researchers worry about audio forgery.
Dear fellow scholars, this is two minute papers with. Earlier we talked about Google's WaveNet, a learning based text to speech engine. This means that we give it a piece of written text and after a training step using someone's voice, it has to read it aloud using this person's voice as convincingly as possible. And this follow up work is about making it even more convincing. Before we go into it, let's marvel at these new results together.
Generative Adversarial network or Variational Autoencoder. He has read the whole thing. He reads books. This is really awesome. This is your personal assistant Google Home. This is your personal assistant Google home. The buses aren't the problem, they actually provide a solution. The buses aren't the problem, they actually provide a solution. She sells seashells on the seashore. The shells she sells are seashells. I'm sure.
As you can hear, it is great at prosody, stress and intonation which leads to really believable human speech. The magic component in the original wavenet paper was introducing dilated convolutions for this problem. This makes large skips in the input data so we have a better global view of it. It is a bit like increasing the receptive field of the eye so we can see the entire landscape and not only a tree on a photograph. The magic component in this new work is using Mel spectrograms as an input to wavenet. This is an intermediate representation that is based on the human perception that records not only how different words should be pronounced, but but the expected volumes and intonations as well. The new model was trained on about 24 hours of speech data and of course no research work should come without some sort of validation. The first is recording the mean opinion scores for previous algorithms, this one and real professional voice recordings. The mean opinion score is a number that describes how a sound sample would pass as genuine human speech. But the new algorithm passed with flying colors. An even more practical evaluation was also done in the form of a user study where people were listening to the synthesized samples and professional voice narrators and had to guess which one is which. And this is truly incredible because most of the time people had no idea which was which. If you don't believe it, we'll try this ourselves in a moment. A very small but statistically significant tendency towards favoring the real footage was recorded. Likely because some words like Merlo are mispronounced automatically voiced audiobooks, automatic voice narration for video games. Bring it on. What a time to be alive. Note that producing these waveforms is not real time and still takes quite a while to progress along that direction. Scientists at DeepMind wrote a heck of a paper where they sped WaveNet up a thousand times. Leave a comment if you would like to hear more about it in a future episode. And of course, new inventions like this will also raise new challenges down the line. It may be that voice recordings will become much easier to forge and be less useful as evidence unless we find new measures to verify their authenticity. For instance, to sign them like we do with software. In closing, a few audio sample pairs. One of them is real, one of them is synthesized. What do you think which is which? Leave a comment below that girl did.
A video about Star wars lipstick that girl did a video about Star wars lipstick she earned a doctorate in sociology at Columbia University. She earned a doctorate in sociology at Columbia University George Washington was the first President of the United States George Washington was the first President of the United States I'm too busy for romance I'm too busy for romance.
I'll just leave a quick hint here that I found on the webpage. Oh, there you go. If you have enjoyed this episode, please make sure to support us on Patreon. This is how we can keep the show running. And you know the drill. $1 is almost nothing, but it keeps the papers coming. Thanks for watching and for your generous support and I'll see you next time. Next time.
