Putting on my tinfoil hat here, if all it takes to build a speech model to impersonate someone's voice is an hour's worth of them talking... what happens when the wrong person gets that? For example, the government or a corporation (a internet phone service, maybe) uses to fabricate evidence of conversations that never really happened; could also be used to aid in identity theft
Microsoft turns spoken English into spoken Mandarin – in the same voice
31–40 of 124 posts
Re: Microsoft turns spoken English into spoken Mandarin – in the same voice
#32The big question is when do we see this in a real product. Many interesting innovations have come out of Microsoft research. It will be nice if they reach end users as well
I can see this going into Windows 8 phones... I'd buy one.
Re: Microsoft turns spoken English into spoken Mandarin – in the same voice
#33Near real time speech to speech translation is awesome[1], but the voice sounded more like how i would picture ASIMO speaking (ie a 1980s speech synthesis) than 'his' voice. 1. is anyone here fluent in mandarin to assess the quality of the output?
Mandarin speaker here. I'm more impressed that it was able to reconstruct the sentences properly (where traditional translation tools typically fail). The output was fine. Obviously doesn't sound like a human speaker, but the tones are correct.
Re: Microsoft turns spoken English into spoken Mandarin – in the same voice
#34Re: Microsoft turns spoken English into spoken Mandarin – in the same voice
#35Re: Microsoft turns spoken English into spoken Mandarin – in the same voice
#36The implications of this kind of technology reaching consumers in the next decade or so are really interesting. If we can get to the point of having handheld devices that can accomplish live translation of spoken word, what exactly is the point of different languages anymore?
I don't follow your logic... Without those different languages, you wouldn't have anything to translate in the first place. If anything, this type of technology promotes independent and different languages, as it makes it so much easier to communicate with others regardless of your native tongue. Also, bravo to Microsoft; I'll remove my jaw from the floor after I watch your video a second time.
different languages emerged because of physical separation. As barriers have been reduced because of technology (both physical and communication barriers) there really aren't "borders" anymore. I can call someone in China right now if I wanted to, something impossible to do even just 100 years ago.
So if we can all talk to each other across the world in real time, and we can all understand each other because of this technology, what exactly is the point of different languages anymore?
Re: Microsoft turns spoken English into spoken Mandarin – in the same voice
#37Now imagine this on Skype as a premium feature.
Re: Microsoft turns spoken English into spoken Mandarin – in the same voice
#38Couldn't this be easily hacked together using google translate and a text to voice program?
Which makes me wonder: is there a text to voice application based on convolution?
Re: Microsoft turns spoken English into spoken Mandarin – in the same voice
#39Earlier quoted context omitted.
Got a Kinect for your XBox yet?
True, but the Kinect is poorly implemented. I find the Kinect incredibly frustrating and usually have it unplugged. When you're watching a show that has a lot of dialogue it will interpret the talking from the show as commands. These commands include things like "Fast forward 40x" or "Rewind 20x".
Interesting -- I've never had that happen. You may want to recalibrate the microphones on your Kinect; it does some crazy analysis to filter audio from the speakers out from the input stream.
Re: Microsoft turns spoken English into spoken Mandarin – in the same voice
#40Earlier quoted context omitted.
> Corporate pages are often translated paragraph by paragraph by humans which results in perfect raw data to train these algorithms. Actually this may be one of the reasons why Google's Japanese translations are so terrible . The why isn't really relevant here[0] (perhaps you already know anyway) but it there are times when the raw data becomes the most misleading. [0] Obviously I still mean those actually translatin…
I speak bad Japanese after having worked there for a while. But Japanese is extremely context dependent. They leave out a lot of words and you just deduce the meaning. (Don't hit me if that's wrong, but that is what I remember thinking back when I was learning it.)
The way you phrase that is a bit misleading. It's not that Japanese speech has any less info than most languages, it's just that you can set topics (add state to a stack, to put it in geek terms) that carry over into subsequent phrases. The correct translation of an individual sentence may thus depend on that previous context.
A simple but famous case is "Watashi wa hamburger desu". The sentence has no subject, so with no context that would get translated as "I am a hamburger", but if you fill in a previously defined subject, it could be "I [order] a hamburger", "My [favorite food] is a hamburger", etc.