Google Chrome is testing a new API that can offer real-time language translation using machine learning. With this tool, users will be able to quickly, easily, and efficiently translate any entered text or an entire web page.
According to Bleeping Computer, the feature is being developed by Chrome's built-in AI team and aims to expand the browser and the ability to download additional language models.
See also: Chrome switches to ML-KEM quantum encryption
Chrome and Edge already have translation features. However, sometimes there are problems with translating web pages that have complex content. For example, Chrome may not be able to correctly translate all sections of an interactive website. Such websites may have their own translation tool, but this could be resource-intensive and slow. With Google's new feature, the process will become faster than ever. Developers will be able to access Google Chrome's translation feature through a simple JavaScript API.
“To perform translation in such cases, websites must, currently, either call cloud APIs or bring in their own translation models and execute them using technologies like WebAssembly and WebGPU,” Google explains.
“This proposal introduces a new JavaScript API for exposing a browser’s existing language translation capabilities to web pages, so that if they exist, they can serve as a simpler and less demanding alternative,” Google explains.
The new AI/API feature will simplify the process by using Chrome's built-in translation engine, for existing content or for imported text, such as a forum or real-time chat service.
Google's feature will allow downloading a machine learning to correctly translate text if the required translation model is not built into the browser.
See also: Google unveils new security features in Chrome
This innovation aims to break down language barriers on the internet, allowing users to more seamlessly access content in their native language. By incorporating advanced machine learning techniques, the API can understand the context, idioms, and nuances of different languages to provide more accurate translations. This is particularly beneficial for businesses looking to expand their global reach, educators providing multilingual resources, and users seeking information that would otherwise be inaccessible due to language restrictions.
Additionally, machine learning models are constantly learning and adapting to improve their accuracy and performance. This means that as more data is fed into the system, translations will continue to become more accurate and natural, making it even easier for individuals to connect with others regardless of language barriers.
However, this feature may bring to light some privacy issues, as it could potentially allow a website to create profiles for a user based on the supported languages and availability status of the browser.
“The most obvious identifier in the current API design is the list of supported languages and particularly their availability status (“none”, “immediately”, or “after download”),” Google warns.
See also: How to translate text in images with Google Translate
“For example, so far, Firefox supports 9 languages, each of which can be downloaded separately. This gives 9 bits of identifying information that websites can associate with each other“.
It's unclear when the new complex text to the Chrome browser, but Google is actively exploring and discussing the idea with the open source community.
Source: www.bleepingcomputer.com
