A RetroSearch Logo

Home - News ( United States | United Kingdom | Italy | Germany ) - Football scores

Search Query:

Showing content from https://www.chrispecoraro.com/microsoft-bing-translator-and-google-translate-compared/index.html below:

Microsoft Bing Translator and Google Translate Compared – Christopher Pecoraro

To compare the differences in quality between Microsoft’s Bing Translator and Google Translate, eight phrases were taken unaltered from Italian tourism websites and each one was automatically translated using both tools. The original phrase is shown first (in the Italian language) along with each translation into English.

Evaluation: Both translations are nicely done, although Bing’s “open also by bike” is a strange sounding-construction. Also, The British English “centre” spelling was chosen even though no region was specified.

Evaluation: Google’s “The B & B in the real alternative“ is not correct although neither is Bing’s “to stay in Taormina historic centre.”

Evaluation: Bing’s “pampering and satisfy your” is grammatically incorrect and Google’s “the service is attentive” sounds strange.

Evaluation: Google gets confused and produces the following grammatically incorrect phrases: “path enlightenment”, “rivers stone”, and “voice of water that burst”.

Evaluation: Strangely, both Google and Bing get confused with “places of historical, cultural and religious”.

Evaluation: Similarly, as in the last example, again both Google and Bing get confused with “green countryside of Corleone invite you”, “agricultural and Corleone Sicily”, and ”corner of the contact with”.

Evaluation: Google in this case got confused with “rooms” which should be only one “room”, while Bing incorrectly writes “elegantly and folklore-inspired art”.

Evaluation: Both Bing and Google incorrectly produce “the specialties of the rooms” and “their special rooms”.

Conclusion: In trying to determine which machine-translation tool is better for short phrases, neither Google nor Bing is the clear winner with the test data that we have selected. Both are incredibly accurate machine translation tools. The accuracy appears to be around 95%; however, a native speaker should always confirm the correctness of the results before using them in both digital and print media. In the case of digital media, the translation should be periodically updated, perhaps in an automated way, since both algorithms are becoming more and more accurate with time.


RetroSearch is an open source project built by @garambo | Open a GitHub Issue

Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo

HTML: 3.2 | Encoding: UTF-8 | Version: 0.7.3