Gboard update with AI handwriting feature

Gboard update with AI handwriting feature

Now Google has added the handwriting recognition feature in its Gboard.

The updated version of Gboard is improved with Machine Learning advancements that allowed Google to come out with new model architectures and training methodologies. This machine learning model operates on the whole input and reduces error rates significantly as compared to the old version. Gboard is now quite fast and makes less mistakes than before.


Android users now can write on the screen. It will be helpful especially for those who are non-English speakers. From hand-designed heuristics, the Handwriting Input now relies on newer architectures and training methodologies. Gboard now supports about 82 languages as Google keeps on adding more languages. The handwriting recognition has been added to Gboard. Over 100 languages are ready so everyone in the world can use Gboard.

According to Google,

Since Gboard is used on a wide variety of devices and screen resolutions our first step is to normalize the touch-point coordinates. Then, in order to capture the shape of the data accurately, we convert the sequence of points into a sequence of cubic Bézier curves to use as inputs to a recurrent neural network (RNN) that is trained to accurately identify the character being written (more on that step below). We will continue to push the envelope beyond improving the Latin-script language recognizers. The Handwriting Team is already hard at work launching new models for all our supported handwriting languages in Gboard.

The Gboard has achieved the lowest latency by converting the recognition models to TensorFlow Lite models. This will make the ultimate user-experience as they are expected to be fast.


You can get more details about the Handwriting Recognition in Gboard on the Google AI blog.


News Source

Recommended for you