Automatic translation technology would solve the problems that arise when trying to translate documents and speeches from one language to another.
As reported by Cellular-news, twenty-three official languages in Europe means more than a billion euros a year are spent to translate and interpret documents or speeches. Automatic translation technology would save time and costs and is currently under development, with eleven partners involved in the program.
The TC-Star project seeks to solve the problem of taking speech in one language and translating it so that it is accurate in another language. Three technologies are used to do so, automatic speech recognition (ASR), spoken language translation (SLT) and text to speech (TTS). ASR transcribes spoken words to text, SLT translates the original language to the language desired and TTS synthesizes the spoken output.
Over time, the accuracy of these systems has improved between forty and sixty percent. Up to seventy percent of words were translated properly. Work is still needed on proper placement in sentences. Recordings of speeches from the European Parliament were used, with both Spanish and English being used. Radio news broadcasts were also used, with Chinese and English being the languages spoken.
It is expected that within a few years, a commercially viable automatic speech-to-speech translator may be available.
To read more, head to the article located here.