There are about 6,500 spoken languages in the world. Google Translate has helped many to decode and understand words, sentences and paragraphs. It is free and fast, although not 100% accurate with larger blocks of text or for some languages.  This is possible because of Statistical Machine Translation (SMT). Computers analyse millions of existing translated documents from the web to learn vocabulary and look for language patterns. After that, Google translates it and picks the most statistically probable translation.

Google Translate

When you start typing in the search box, it will anticipate what you’re looking for, then provide suggested search terms for the same based on past searches, trends, or from your present location. For example, bus traffic delays. Google gets all the real-time data on bus locations and forecasts, and can now provide predictions so you don’t have to rely on bus schedules.

Google Search and Google Maps

Google’s main aim is to organise the world’s information and make it accessible and useful to everyone. This is where AI comes in. Google has mastered machine learning and made life easy with all its applications. It is continuously updating its platform and is one step ahead. Google applies AI to products and new domains and helps to develop tools to ensure that everyone can access AI.

Google is a machine learning-first company

Google Adsense. It keeps track of the users’ search history and recommends advertisements to users as it is now aware of its target market. With Google Adsense, Adsense matches text and display ads to the site based on the content and the visitors, while the website owners earn money from their online content. Google Adwords. Google Ads help advertisers to pay to display short advertisements.

Google Adsense and Google Adwords

Google Photos play an essential role for Gen X and Gen Z. The app stores your gallery online and lets you access it anytime and anywhere. It also automatically collects and organises pictures of the same people or objects into albums, such as on a vacation, office, etc., then shows each album's ‘best’ photos. The photos are tagged with well-known landmarks they depict and people that appear frequently. It tracks down space-wasting documents.

Google Photos

Machine learning algorithms are trained with examples, where they learn from past experiences and analyse historical data. As it continues training, it can identify patterns to make predictions about the future. Google takes advantage of machine learning algorithms and systematically apply them across all their products, such as search, ads, YouTube, or Play. Machine learning is already part of services like Gmail, Google Search and Google Maps.

How Google uses machine learning

Social, promotional and primary mails are filtered through Google as it labels the email accordingly. When a user marks a message in a consistent direction, Gmail will learn your results for categorisation for the future and change the uses. With the help of smart replies, you can reply within seconds. Gmail will suggest a ‘Smart Reply’ and ‘Smart Compose’.

Gmail

Speech Recognition. This feature enables a user to convert audio to text by applying neural network models in an easy to use API. The API currently recognises 120 languages and their variants. Reverse Image Search. Image search creates categories you might want. With the image search, it can be easy to search for similar images and the websites that contain these images.

Speech Recognition and Reverse Image Search

Google Assistant. It helps with everyday tasks. For example, it makes it easy to search for nearby restaurants, buy movie tickets while on the go and find the nearest theatre to you. It can also help to navigate to the theatre. Google Play Music have 30 million-plus songs available for streaming. It can handle up to 50,000 own songs without any subscription.

Google Assistant and Google Music

Deep-dive into google's machine learning models and how it relies more on logic than data