How many times have you used Google Translate this week? People use Google Translate all around the world daily, using up to 100 billion words. Not only people rely on Google Translate but businesses too. Google Translation enables organizations to dynamically translate between languages using Google’s pre-trained or custom machine learning models.

Google Cloud Translation API

Initially available as a web app to detect and translate between languages, Google Translate is now also available as an API. Google Translate API supports more than one hundred different languages, from Afrikaans to Zulu. Used in combination, this enables translation between thousands of language pairs. Google Cloud Translation API is available in Basic and Advanced package. Advanced Translation API offers dynamic results just as quickly as the Basic, but also includes other customization features, which is very important when you use phrases or terms that are specific in areas or in contexts. Google Cloud Translation API is suitable for companies of any size.

Google AutoML Translation

Google AutoML Translation is a service provided by Google to create custom translation models when Google Translation API is not enough because you have specific vocabulary. AutoML Translation can help you automate the translation job in a scalable way, allowing you to enter new markets quickly.

AutoML Translation dynamically detects and translates between languages, translate with custom models, while supports 50 language pairs. Developers can build their own custom machine learning model in minutes because cloud AutoML enables developers with limited machine learning expertise to train high-quality models specific to their business needs. Google AutoML Translation is simple to use.  You create Dataset from your custom translations and Google creates a custom model from the generic Translation API model and adds a layer that specifically helps the model get the right translation based on your Dataset. AutoML Translate enables you to perform supervised learning, which involves training a computer to recognize patterns from translated sentence pairs.

Google Cloud Translation API vs Google AutoML Translation

The Translation API covers a huge number of language pairs and does a great job with general-purpose text. Where AutoML Translation really shines is for the "last mile" between generic translation tasks and specific, niche vocabularies. Google Translation custom models start from the generic Translation API model but add a layer that specifically helps the model get the right translation for domain-specific content that matters to you. Translation charges on a per-character basis, even if the character is multiple bytes. AutoML Translation charges for training, per-character prediction usage, and storage. You only pay for what you use, so Google’s products are usually more affordable than other options.

Google Media Translation API

Media Translation API delivers real-time speech translation to your content and applications directly from your audio data, also enhances the accuracy of interpretation by optimizing model integrations from audio to text and abstracts potential frictions you may face initiating multiple API calls. Media Translation API provides real-time translation during streaming audio input from a microphone or prerecorded audio files. Media Translation supports streaming translation of speech in one language (the source language) to text in another language (the target language). It is optimized to recognize and translate from US English as the source language with another target language, but several other languages are supported as the source language. All together Media Translation API supports 12 languages.

Softline’s Google expert team has a huge international experience in Google Solutions, Google Translation included.  Softline experts provide you full service so your company can utilize all the benefits of Google Translation solutions.