14 DAY TRIAL // The ability to hear the translations in Google Translated spoken with the proper enunciation, even if with a slightly creepy, robotic voice, is one of the most useful tools of the product. And now, the creepy voice - voices actually - are getting better with a roll out of improved synthesized voices for many of the languages supported in Translate.
"One of the features of Google Translate that users enjoy the most is the ability to listen to the text they have just translated in audio form," Alex Salcianu, Software Engineer at Google, wrote.
To play an audio version of the translated text we use a speech synthesizer, a computer algorithm that converts text to speech. Today, we have launched new speech synthesizer voices," he announced.
Google is introducing support for three new voices with this roll out. Arabic, Japanese and Korea are now supported for the first time.
But the existing ones have not been neglected. The voices for 17 languages have now been improved, specifically: Czech, Chinese, Danish, Dutch, Finnish, French, German, Greek, Hungarian, Italian, Norwegian, Polish, Portuguese, Russian, Spanish, Swedish and Turkish.
If you've been using the voice feature in Google Translate, the differences should be noticeable. Of course, the quality of voice synthesis has constantly improved for Google Translate. It's not perfect, but it does sound significantly better than the robotic voices initially introduced.
It's interesting that Google very recently acquired Phonetic Arts, a British company that just so happens to specialize in synthetic speech. At the time, Google said that the filed of synthetic speech had a lot of applications and that the company was very interested in the technology.
While it may be too soon for Google to already be integrating Phonetic Arts' products into it's own, it does show that the company has bigger plans for this type of technology and they go beyond Google Translate.