Hello everyone, my name is Alex and I’m the Founder of Translation Cloud. We’ve just posted a video to YouTube on how to translate your website with the Google Translate API.
Before watching this video or reading the rest of this post, please check out our last three posts in this series for a great overview of the biggest players in the field of machine translation:
Google was one of the first major companies to look into machine translation as a viable source of revenue. Their search engine was innovative, and brought a totally new idea of exchanging links and measuring link quality in terms of how to rank websites in their search engine. They also looked at the languages of the web, and around 2007 they released their own machine translator, a simple text field for copying and pasting text in the browser where the machine would do the translation work for you. Later on, Google developed their translation API for programmers to integrate into their applications. Google would work in the background, parse and translate the results in real time, and provide users with a true multilingual experience.
When we started ConveyThis back in 2008, it was a smash hit when we incorporated Google as one of the machine translators. Then, Google offered only 53 languages. But fast-forward ten years and Google now supports over 100 languages, and translates them all fairly well.
In the beginning, Google’s API was free, supported 53 languages, and did so-so work. But today, it has a neural capacity–in essence, artificial intelligence to learn from its experiences and interactions. Therefore, Google has been able to boast that their translations have been indistinguishable from those of an average human.
Linguists complain that humans are irreplaceable, and that someone should still be overlooking the translations, and this is true. A new industry of “post editing” has popped up, a hybrid of machine translation proofread by human translators, which brought down the prices of translation for certain languages and reduced the turnaround time for projects.
Google is understandably the leader in the market, and their API is one of the most expensive ones on the market. They know their brand and quality, and they are twice as expensive as Microsoft per 1 million characters translated, and $5 more than the Yandex Translate API. It supports more languages, and translates more languages in a neural network capacity than Microsoft. It also can be easily integrated with other Google APIs (such as text to speech and speech recognition APIs).
There is no free tier whatsoever. Even though Google started as a free API, as the quality grew and Google became more and more involved in the project, they removed the free tier completely. If you need something to experiment with, this will not be the best solution for you.
If you learned something from this video, please give it a like and subscribe to our channel!