PhoneArena is looking for new authors! To view all available positions, click here.

Phone Links

more » more »
  • Home
  • ›
  • News
  • ›
  • Microsoft makes strides in direct speech translation, we hope to see it in Windows Phones (video)

Microsoft makes strides in direct speech translation, we hope to see it in Windows Phones (video)

Posted: , by Daniel P.

Categories:

Share:

Discuss 6
Microsoft makes strides in direct speech translation, we hope to see it in Windows Phones (video)
The speech translation technology of Microsoft has been in the labs for quite a while, and is apparently getting ready for prime time, as Redmond decided to demonstrate what it can do.

It is essentially on-the-fly converting of your words to a different language, and with your own voice to boot. It is not perfect yet, and gets one in eight words wrong on average, but just think about never having to learn a foreign language, and still practice your best sarcastic remarks on the locals when they serve you food that still looks like it's moving.

The software has to be provided first with hours of recordings, so as it can analyze the subtleties of your vocal cords, and then learn to speak Cantonese with them, just like the first voice recognition efforts required you to talk to the computer for quite a while until it gets your brand of crazy accent. Rick Rashid, the head of Microsoft Research who demonstrated the technology, clarified how this became possible:

Just over two years ago, researchers at Microsoft Research and the University of Toronto made another breakthrough. By using a technique called Deep Neural Networks, which is patterned after human brain behavior, researchers were able to train more discriminative and better speech recognizers than previous methods.

[...] We have been able to reduce the word error rate for speech by over 30% compared to previous methods. This means that rather than having one word in 4 or 5 incorrect, now the error rate is one word in 7 or 8.


If anyone is still wondering why do we need desktop-class processors in our phones and tablets, it's features like these that will evidently require all the horsepower your Windows Phone can muster, and we hope this direct translation tech will arrive soon to our handsets. The fun starts about 5:30 into the video below, but it is worth watching in its entirety. 

via TNW

Share:

Discuss6

6 Comments

  • Options
    Close




Want to comment? Please login or register.

1. drahmad posted on 09 Nov 2012, 03:12 5

thats amazing....this is innovation.

Reply

2. jhp2025 posted on 09 Nov 2012, 03:19

this is what I've been waiting for since the first StarWars movie I've ever seen! no time to wait for it Bill...!

Reply

3. nbringer posted on 09 Nov 2012, 03:29

"when they serve you food that still looks like it's moving" - or looks like a hamburger...

Reply

4. rsxl5 posted on 09 Nov 2012, 05:06

XXI century breakthrough announcement 480p....

Well, I'm not native speaker of Chinese but I do speak Chinese for last 9 years and we still have very long way to go (despite the fact that Chinese grammatic is very in-complex (not like ie: French or Polish one),

From what I see part of the translation is not exactly the same (or as shown in a sentence below it's incorrect (7m23s)). It's understandable for me in both languages but due to the fact that I did a few languages (while younger) I always look for closest match to make sens. It's still word-copy translation thou.

ON VIDEO:
- 'I'm speaking in English and and hopefully you will hear me speaking Chinese in my own voice'
- '我是说英语,希望你能听我说中国人自​己的声音。'

Word 'Chinese' from English sentence was translated as a Chinese Person (中国人) instead of Chinese Language (华语 or other word describing language (they have a few of them)).

(IMO) Proper translation would be rather:
我是说英语，希望你会听到我说华语

Reply

5. drahmad posted on 09 Nov 2012, 06:43

he already said we still have to go a long way. so......

Reply

6. fervid posted on 09 Nov 2012, 08:40

That sounds more like a where is it used context. Chinese could mean the language or the people. Much like other things in the English language that speach applications have issues with- Like Dr. - Does that mean Drive like a street or Doctor like the person. Then you have words like read that can be pronounced two different ways. It is going to be difficult for a computer to handle every case. Much like teaching a 3yr old to speak and why things are said certain ways.

Reply

Want to comment? Please login or register.

Hot Phones

  • Samsung Galaxy S4Samsung Galaxy S4
  • Apple iPhone 5Apple iPhone 5
  • Samsung GALAXY Note IISamsung GALAXY Note II
  • HTC OneHTC One
  • Sony Xperia ZSony Xperia Z
  • Google Nexus 4Google Nexus 4
  • LG Optimus GLG Optimus G
  • Nokia Lumia 920Nokia Lumia 920