Improved Accuracy, Faster Speeds, & New Languages for Watson Language Translator
6,500
The number of languages used around the globe. When you look around, on the web or in daily life, you begin to notice the importance of translation. Companies around the world leverage the power of translation in their daily processes to make information accessible. Living in the information age means websites, documents, books, chatbots all need to be available for anyone to access at a moment’s notice. Given the amount of information, there are not enough people to help bridge the language barrier without the assistance of machine translation.
Translation is one of the hardest challenges to tackle, but it is the most promising in the field of Artificial Intelligence.
Neural Machine Translation Built In
From rule-based machine translation to statistical machine translation there have been many previous approaches to machine translation. Today, by default, all new Watson Language Translator users will gain access to our Neural Machine Translation (NMT) technology. With this update you can expect to see improved speed and accuracy.
NMT is the latest translation method based on Deep Learning that has led to remarkable improvements in translation fluency, achieving higher human evaluations compared to previous technologies.
By accessing the new version of the API, your translation request will use the NMT models for the following languages:
Arabic, Bulgarian, Catalan, Chinese (Simplified & Traditional), Croatian, Czech, Danish, Dutch, English, Estonian, Finnish, French, German, Greek, Hebrew, Hindi, Hungarian, Irish (Gaelic), Italian, Indonesian, Japanese, Korean, Latvian, Lithuanian, Malay, Norwegian, Polish, Portuguese (Brazil), Romanian, Russian, Slovak, Slovenian, Spanish, Swedish, Thai, Turkish, Urdu and Vietnamese
Existing User?
If you made your account before June 15th, you will need to migrate to v3 before October 2018 to ensure all functionality remains.
Check out our Migration Guide
Improve your translations!
Get started with the new Watson Language Translator in 2 simple steps.
Step 1: Creating Your Account
To access the API you will need to have an IBM Cloud Account. Once you provision an instance of Watson LT, take note of your service credentials.
Step 2: Translating Text
Once you have your username and password, you can get started by calling the API from the command line. Let’s translate two sentences from French to English.
Sample Input (Mac):
curl -X POST \
--user username:password \
--header "Content-Type: application/json" \
--data '{"text":"Hello, world","source":"en","target":"fr"}' \
https://gateway.watsonplatform.net/language-translator/api/v3/translate?version=2018-05-01Sample Output:
{
"translations": [
{
"translation": "Bonjour, le monde!"
}
],
"word_count": 2,
"character_count": 13
}The output will give you the translation, word count, and character count. For more ways to call the API, check out our API reference guide.
Additional Links:
Product Homepage | v3 Release Notes | Getting Started | Sample Apps
Zia Mohammad is on the product management team for Watson Language Translator and Watson Natural Language Classifier. His passion centers around: AI, emerging technologies, space, and neuroengineering.
