
Google’s continuing to push technological boundaries. In language, Google has been working on both text translation and voice recognition. It only seems logical to combine the two for real-time speech translation. And even better yet, real-time speech translation built into the software of a smartphone.
“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.”
It’s the smartphone that Google hopes will aid this quest. Because smartphones are personal, it’s possible to have them learn the speech patterns and intonations of the user. The software doesn’t have to master all language possibilities, just yours. This would simplify the task of providing a machined-based translation.
While Google’s idea is interesting, not all are convinced a solution is right around the corner. David Crystal, an honorary professor of linguistics at Bangor University, says, “The problem with speech recognition is the variability in accents. No system at the moment can handle that properly.” Crystal says its possible Google will get their first, but it will probably take a few years longer than than expected.

































