Fichier de travail (INPUT) : ./DUMP-TEXT/2-9.txt

Encodage utilisé (INPUT) : utf-8

Forme recherchée : translation|traduction|机器

• Ligne n°143 : Machine translation—is it good enough for customer support?


• Ligne n°147 : In 1954, at the first public demonstration of a machine translation
Ligne n°148 : (MT) system, researchers from IBM declared that translation would be ...

Ligne n°147 : ... In 1954, at the first public demonstration of a machine translation
• Ligne n°148 : (MT) system, researchers from IBM declared that translation would be
Ligne n°149 : fully handled by machines within the next few years. ...


• Ligne n°155 : "One naturally wonders if the problem of translation could conceivably
Ligne n°156 : be treated as a problem in cryptography. When I look at an article in ...

Ligne n°160 : ... A few years later, Weaver followed up with this: "No reasonable person
• Ligne n°161 : thinks that a machine translation can ever achieve elegance and style."

Ligne n°166 : ... service, the output might look a lot like you're decoding secret
• Ligne n°167 : messages. It raises the question: Why is translation so difficult?

Ligne n°170 : ... researchers in the field, explained over lunch, "Humans have a huge
• Ligne n°171 : knowledge about the world and that is crucial for translation. As a
Ligne n°172 : human, I can understand the context, the culture of a specific ...

Ligne n°172 : ... human, I can understand the context, the culture of a specific
• Ligne n°173 : language, and that is key in translation. Machines don’t have yet that
Ligne n°174 : kind of knowledge." ...

Ligne n°176 : ... As a human, I can understand the context, the culture of a specific
• Ligne n°177 : language, and that is key in translation. -André Martins


• Ligne n°193 : Machine + human translation in customer support

Ligne n°195 : ... For most people, Google Translate is the first thing that comes to mind
• Ligne n°196 : at the mention of machine translation. But would you trust Google to
Ligne n°197 : translate everything you need to tell your customers? ...

Ligne n°199 : ... Probably not—for this or that reason (and many others). The thing is, a
• Ligne n°200 : human quality translation without grammar mistakes, typos, and
Ligne n°201 : mistranslations requires a human. And the tech giant is not alone in ...

Ligne n°200 : ... human quality translation without grammar mistakes, typos, and
• Ligne n°201 : mistranslations requires a human. And the tech giant is not alone in
Ligne n°202 : facing these challenges when it comes to NMT—nor is it the only company ...

Ligne n°206 : ... one of the fathers of MT and a professor at the ADAPT Center at Dublin
• Ligne n°207 : City University. Why? Because for neural machine translation to work we
Ligne n°208 : need human translations to feed into the systems and train them. Once ...

Ligne n°207 : ... City University. Why? Because for neural machine translation to work we
• Ligne n°208 : need human translations to feed into the systems and train them. Once
Ligne n°209 : the system receives all the data, it starts to learn patterns and to ...

Ligne n°209 : ... the system receives all the data, it starts to learn patterns and to
• Ligne n°210 : produce better translations. And today you can’t do that with Google.


• Ligne n°212 : Neural machine translation relies on human translations that feed into


• Ligne n°212 : Neural machine translation relies on human translations that feed into
Ligne n°213 : the systems and trains them. ...

Ligne n°215 : ... So, does this mean you can reliably translate customer support emails
• Ligne n°216 : or even real-time chats using machine translation? Has the technology
Ligne n°217 : reached its full potential? ...


• Ligne n°219 : Pure machine translation systems lack the ability to adapt to different
Ligne n°220 : content types (FAQs vs. customer tickets, for instance), and they are ...


• Ligne n°224 : In short, pure machine translation systems lack the "human touch"
Ligne n°225 : required for understanding cultural references and contextual ...

Ligne n°226 : ... differences. Today, however, NMT combined with advanced, automated
• Ligne n°227 : quality assurance and post-editing by humans, ensures translations that
Ligne n°228 : are sound, and sound good—and often delivered within 20 minutes. ...

Ligne n°228 : ... are sound, and sound good—and often delivered within 20 minutes.
• Ligne n°229 : Machine translation may not be at the end of its road, but it’s come a
Ligne n°230 : long way toward meeting critical business needs—allowing companies to ...


• Ligne n°234 : By itself, machine translation will always sound like a "better parrot"
Ligne n°235 : but aided by data and people, it can offer an elegant solution to ...