AM (Cross-posted from the Google Translate Blog)
One of the popular features of Google Translate is the ability to hear translations spoken out loud (”text-to-speech”) by clicking the speaker icon beside some translations, like the one below.
We rolled this feature out for English and Haitian Creole translations a few months ago and added French, Italian, German, Hindi and Spanish a couple of weeks ago. Now we’re bringing text-to-speech to even more languages with the open source speech synthesizer, eSpeak.
By integrating eSpeak we’re adding text-to-speech functionality for Afrikaans, Albanian, Catalan, Chinese (Mandarin), Croatian, Czech, Danish, Dutch, Finnish, Greek, Hungarian, Icelandic, Indonesian, Latvian, Macedonian, Norwegian, Polish, Portuguese, Romanian, Russian, Serbian, Slovak, Swahili, Swedish, Turkish, Vietnamese and Welsh.
You may notice that the audio quality of these languages isn’t at the same level as the previously released languages. Clear and accurate speech technology is difficult to perfect, but we will continue to improve the performance and number of languages that are supported.
Posted by Fergus Henderson, Software Engineer – googleblog