09 February 2010 ~ 4 Comments

Google developing real-time spoken language translation service

Image credit: voices.com

In a project that could revolutionize international communication, Google is working to create mobile phone software that will build on voice recognition and automatic translation technology to provide almost instantaneous voice translations. Google says the project should be ready in a couple of years. The UK Times reports:

“We think speech-to-speech translation should be possible and work reasonably well in a few years’ time,” said Franz Och, Google’s head of translation services.

“Clearly, for it to work smoothly, you need a combination of high-accuracy machine translation and high-accuracy voice recognition, and that’s what we’re working on.

“If you look at the progress in machine translation and corresponding advances in voice recognition, there has been huge progress recently.”

Although automatic text translators are now reasonably effective, voice recognition has proved more challenging.

“Everyone has a different voice, accent and pitch,” said Och. “But recognition should be effective with mobile phones because by nature they are personal to you. The phone should get a feel for your voice from past voice search queries, for example.”

While considerable hurdles remain in creating effective voice translation, this project promises to create excitement as Google works to create a well-functioning product. Read more from the Times Online.

4 Responses to “Google developing real-time spoken language translation service”

  1. Jeff 10 February 2010 at 3:17 am Permalink

    That is a real game changer.

  2. Don Johnson 15 February 2010 at 6:43 pm Permalink

    Equating written text to human speech is a huge error. Even with written text, the current translation services are laughable. Let alone voice to text. I have a google voice account, which I love, but the voice-to-text sucks, and is, at best, barely understandable.

    This is a much harder nut to crack than anyone thinks. This only works on dedicated devices, like phones, because of the very limited relevance domain. Ask your phone if it likes puppies. It doesn’t respond.

    Also, the human voice is musical. It goes up and down in fourths and fifths (natural intervals) Its happy in majors and sad in minors. I bet nobody is working on translating that. This is one reason written poetry doesn’t sell, but people are riveted at live performances of it (including hip-hop).

    The music in human voices, which is a great wellspring of sub-text and meaning, is looked at as noise to filter out by these current technologies. Tape a pleasurable dinner conversation with good friends and family. Have it transcribed. See how much was lost, even in a perfect transcription.

  3. limewire 16 May 2010 at 9:53 am Permalink

    limewire pro download

  4. limewire basic 27 May 2010 at 3:44 pm Permalink

    Jesu Christ, consultants feuding it out? This is painful. Worse on Mo, but matches the Mculiffe campign pretty we


Leave a Reply