Impossible Real Time Translation Analysis
The Quality of Machine Translation English into Dutch
Language barrier, a thing of the past? That’s the theory of this recent Wall Street journal article, which states that ten years from now we’ll have a near-perfect speech simultaneous machine translation that will also recreate the voice of the speaker in their native language. Author Alec Ross argues that it will bring the world close together.
Machine translation right now doesn’t give you the best translation you need. Everyone knows this. It works OK for simple sentences in similar language. But try using it in a complex sentence like Chinese and English and you’ll get a result that is typically nonsense.
It’s not as simple as just pressing the translate button.
Impossible Real Time Translation. You need to know that languages are structured differently. …show more content…
What I’m saying is this: Computer does not have the ability in translating a language accurately. A human interpreter that is skilled in translating can use all kinds of information to inform her translation, including:
• Personality of the speaker.
• Remembering what has been said at the previous conference.
• Culture and history of the speaker that might help them inform their thinking and phrasing.
• The accent of the speaker and on how they might change pronunciation of some words and phrases.
• Determines the overall mood of the room whether or not something was intended as sarcasm or a joke.
• The audience for the speech
• The gestures that the speaker is making.
• Et cetera
It’s a complex process that requires the intelligent application of the background knowledge about both languages and cultures and the current and recent input of the speech, as well as the ability to read people for their intentions. It is work that requires intelligent analysis, and that means that without artificial intelligence, machine can only go so far.
Machine will need to be pretty human in order to get human-quality
Language barrier, a thing of the past? That’s the theory of this recent Wall Street journal article, which states that ten years from now we’ll have a near-perfect speech simultaneous machine translation that will also recreate the voice of the speaker in their native language. Author Alec Ross argues that it will bring the world close together.
Machine translation right now doesn’t give you the best translation you need. Everyone knows this. It works OK for simple sentences in similar language. But try using it in a complex sentence like Chinese and English and you’ll get a result that is typically nonsense.
It’s not as simple as just pressing the translate button.
Impossible Real Time Translation. You need to know that languages are structured differently. …show more content…
What I’m saying is this: Computer does not have the ability in translating a language accurately. A human interpreter that is skilled in translating can use all kinds of information to inform her translation, including:
• Personality of the speaker.
• Remembering what has been said at the previous conference.
• Culture and history of the speaker that might help them inform their thinking and phrasing.
• The accent of the speaker and on how they might change pronunciation of some words and phrases.
• Determines the overall mood of the room whether or not something was intended as sarcasm or a joke.
• The audience for the speech
• The gestures that the speaker is making.
• Et cetera
It’s a complex process that requires the intelligent application of the background knowledge about both languages and cultures and the current and recent input of the speech, as well as the ability to read people for their intentions. It is work that requires intelligent analysis, and that means that without artificial intelligence, machine can only go so far.
Machine will need to be pretty human in order to get human-quality