191–200 of 311 posts
The samples all sound a little like Rich Little and Stephen Hawking's love child doing impressions: they won't fool very many people. But, you can certainly see where this is going and that's the worrisome part.
Funny thing is, this is approximately where CIA was with similar technology in closer to 2000. They did some demos for politicians about how they can given anyone's fake their messages. That stuff is golden for propaganda means, and for confusing stuff like military chains of command. Today the CIA probably has worked out all the robotic artifacts already, and their output is really indistinguishable.
Earlier quoted context omitted.
Try listening to the NOAA weather radio. They've been using very good TTS for years.
Wow, they are. Very surprised. This is so good I'm wondering whether this is actually a massive (massive) sound bank. I think it might be a sound bank. A random radio receiver site I found that doesn't require Flash, tuned to NOAA for Akron OH: http://tunein.com/radio/NOAA-Weather-Radio-1624-s88289/ The list I got the above link from (^F "noaa"): http://tunein.com/radio/Weather-c100001531/ I suspect the warbling I'm…
The samples weren't that convincing to me, but could probably be used to switch a word here and there. That may be enough.
How difficult is it to create/tune voices from parameters rather than training from an audio clip? I build software where people create fictional characters for writing, and having an author "create" voices for each character would be an amazing way to autogenerate audiobooks with their voices, or interact with those characters by voice, or just hear things written from their point of view in their voice for that extra immersion. Having an author upload voice clips of themselves mimicking what they think that character should sound like, but probably would keep traces of their original voice (and feel "fake" to them because they can recognize their own voice), no?
Can't wait to see how this pans out. Signed up for the beta and will definitely be pushing it to its limits when it's ready. :)
This is pretty basic at the moment and it's terrifying. Yeah, it has an MS Sam feel to it, but as the tech improves and we know it will, you could use a service like this to put words in someone's mouth. Think about how you could trip up a CEO or a Politician by playing some random clip that they never said. When that gets into the Zeitgeist judgments will be made in the court of public opinion devoid of facts or rea…
There are human impersonators already. I suppose it's not that easy to fake a visible, high-ranking person for long.
For example, as an individual with hearing problems, I may not be so easily able to determine a synthesized recording from an actual recording - for a short period of time. With longer recordings it may become more obvious.
Earlier quoted context omitted.
Wouldn't you need tons and tons of training data ? Obama, Clinton and Trump are public personalities, so it's easy to have many hours of recording of their voices. A random relative; not so much.
They claim to only need 1 minute of recording. In this age of all the kids sharing everything all the time that shouldn't be too hard to acquire.
Earlier quoted context omitted.
I wouldn't be comfortable watching a movie scene if I knew I was looking at computer-generated faces and voices.
Did you feel that way when seeing Grand Moff Tarkin on "Rogue One"? https://www.wired.com/video/2017/02/how-rogue-one-recreated-...
Funny thing is, this is approximately where CIA was with similar technology in closer to 2000. They did some demos for politicians about how they can given anyone's fake their messages. That stuff is golden for propaganda means, and for confusing stuff like military chains of command. Today the CIA probably has worked out all the robotic artifacts already, and their output is really indistinguishable.
NNets just recently got really good. You are correct though, politicians would love this. I believe technology would make a Judge's life really hard.