Introducing new assistive features in G Suite to help businesses focus on what matters

We’re dedicated to creating products that are helpful for our users, including those in a business setting. G Suite helps companies of all sizes transform how they work by encouraging people to focus on what really matters, whether that be fostering a culture of collaboration like Iron Mountain or reducing IT and infrastructure costs like Nielsen

Today, we’re announcing new ways that Google’s leading artificial intelligence (AI) can help people accomplish more in the workplace. First, we’re adding more assistive features in Google Docs to help you create high-quality work more quickly. In addition, the Google Assistant has been helping people get things done in their personal lives for the past few years, and today, we’re bringing more Assistant features to G Suite users to help them work more efficiently and seamlessly. 

Even more assistive writing features in Google Docs
We’re adding additional AI-powered capabilities within Docs to make it faster and easier for people to create high-quality, error-free content. 

  • Write faster. We’re bringing Smart Compose to Google Docs. Last year, we introduced the feature in Gmail, which uses Google AI to suggest complete sentences as emails are drafted. Smart Compose has already saved people from typing more than 2 billion characters each week. To help people create documents quickly, we’re bringing this same intelligence to Docs. Sign up for the beta.

  • Cut back on grammatical and spelling errors. To build on the intelligence we introduced earlier this year, Docs will start using cutting-edge, neural network technology to power grammar suggestions in the coming weeks. The neural network approach has proven to help individuals catch even more errors in internal experiments (Read more about the research behind this new approach in this post.). In addition, we’re soon bringing spelling autocorrect to Docs. Using the power of Google Search, this intelligent feature constantly learns new words or phrases that become part of the English language from search to inform spelling suggestions in documents, just as it does already in Gmail.

  • Receive suggestions that are tailored to your business. Lastly, we’re extending our machine-learning based spelling suggestions to recognize commonly used words from your business domain to inform recommendations. For example, if your organization has an internal project name or acronym that’s commonly used, Docs will not only stop underlining such terms, but will also suggest corrections when these terms are misspelled. 

Expanding the power of the Google Assistant for G Suite Businesses
Now available in beta, you can accomplish more on the go with the Google Assistant when you’re logged into your G Suite account. We’ve also enabled the Asus Hangouts Meet Hardware kit to respond to voice commands in meeting rooms.

  • Manage your calendar while on the move. Whether you’re on your commute home or on the go between meetings, the Google Assistant can help you manage scheduling in Google Calendar. You can ask the Assistant to read your calendar, create events, cancel events, or even reschedule events. To try this expanded functionality of the Google Assistant for G Suite users within your organization, sign up for the beta

  • Send quick messages and dial into calls hands-free. Beyond managing your calendar, we know that sometimes you need to send messages and join calls on the go. You can say things like, “Hey Google, join my next meeting” or “Hey Google, send an email to my next meeting”. This functionality is also available as part of the beta.  

  • Have a more seamless meeting experience. We’re also making the Google Assistant available in meeting rooms with Asus Hangouts Meet Hardware kit. Instead of clicking into meetings to join them, people can say voice commands to the Assistant to join a meeting, exit a meeting, or even make a phone call. This functionality is available in the Hangouts Meet hardware with the Google Assistant beta today.

  • Increase accessibility in meeting rooms. Another goal in bringing the Google Assistant into meeting rooms is to increase accessibility. As part of the beta, people can say a voice command, like “Hey, Google, turn on spoken feedback,” to use accessibility features without having to find the button to turn them on. Read more about spoken feedback in Hangouts Meet hardware.

We all could use some help at work. These new assistive features, along with capabilities in the Assistant, can help G Suite users make it through their work day with ease. Learn more.

Using neural machine translation to correct grammatical faux pas in Google Docs

First impressions are everything in the workplace, and these often take place in the documents or presentations that we share with others. Spelling or grammatical errors can be distracting and make a proposal look unprofessional—something we all want to avoid. We’re focused on providing more assistive writing capabilities in G Suite to help you put your best work forward, which is why earlier this year we introduced new grammar correction tools in Google Docs to help people write more quickly and accurately. With the help of machine learning, already more than 100 million grammar suggestions are flagged each week.

Advancing grammar suggestions using neural machine translation
To date, Google’s grammar correction system uses machine translation technology. Essentially each suggestion is treated like a translation task–in this case, translating from the language of ‘incorrect grammar’ to the language of ‘correct grammar.’ At a basic level, machine translation performs substitution and reorders words from a source language to a target language, for example, substituting a “source” word in English (“hello”) for a “target” word in Spanish (“hola”). 

With the latest advancements from our research team in the area of language understanding–made possible by neural machine translation–soon, we’re making a significant improvement to how we correct language errors by using Neural Grammar Correction in Docs.

Neural Grammar (in device).gif

How it works
Since Grammatical Error Correction (GEC) can be viewed as “translation” from ungrammatical to grammatical sentences, sequence-to-sequence models developed for neural machine translation can be applied to this task. To train high quality models, we generally want to have millions or billions of examples of parallel data, where each training example consists of a sentence in the source language paired with its translation in the target language. Unlike several other machine translation tasks (such as translating from English to French), there is very little parallel data for GEC. To overcome this challenge, we developed two contrasting methods to generate large quantities of parallel data for GEC:

  • The first method takes good sentences and makes them worse by automatically translating them to some other language and then back to English.

  • The second method extracts source-target pairs from Wikipedia edit histories with a minimal amount of filtration.

You can read more about GEC and some of our approaches in this paper.

To ensure that the models were feasible to deploy on Google Docs without using an unreasonable amount of computing resources, we used Tensor Processing Units (TPUs). TPUs have provided substantial performance increases for many other Google products, including Smart Compose in Gmail. In addition, we used Google’s open source Lingvo TensorFlow library, which enabled us to easily experiment with modeling changes, and also allowed us to carefully optimize how the TPU cores generate suggestions.

What this means for writers
So what does it all mean for you? Well, by applying neural machine translation models to grammar correction, we are able to correct many more of the grammar mistakes you may make while writing. To launch these improvements, we did a lot of testing to ensure that the changes actually are more helpful. Here are some of the examples from our evaluation process that demonstrate neural grammar correction’s capabilities:

What tense is it anyway?

Grammar_03.png

Is it steak or stake?

Grammar_02.png

Changing to the neural machine translation method has shown a marked increase in the recall of grammar correction suggestions in Docs. We hope this update can continue to help you write with ease.

Made in the shade: How Nubian Skin makes more inclusive fashion with G Suite

“Nude” comes in all shapes, sizes and colors. But for Ade Hassan, finding undergarments to match her skin tone took more effort than necessary. Born in the U.K. to Nigerian parents, Hassan searched unsuccessfully for underwear and swimwear that matched her skin—a problem that continued to frustrate her through college and into her professional life. 

“My nude isn’t the nude I see in shops,” Hassan says. 

Like many other black entrepreneurs, she created her own solution. Hassan founded Nubian Skin, a company that sells lingerie, hosiery, and swimwear for women of color. With shades like “cafe au lait” and “berry,” Nubian Skin products reflect diverse skin tones and are sold in shops located in the U.K., Nigeria and the U.S., as well as online. Beyoncé and her dancers even asked Nubian Skin to outfit them for their  2017 “Formation” tour. 

Since nearly the beginning, Hassan has used G Suite to keep her business humming. From managing meetings with retailers and manufacturers, to coordinating publicity efforts with fashion press, to keeping track of retail invoices, G Suite has helped Hassan scale her business as quickly as people have demanded her products. “There are so many pitfalls and so many things that are hard about business,” Hassan says. “It’s really amazing to have something like G Suite that is easy and intuitive.”

Nubian Skin’s fashion fame arrived seemingly overnight. When Hassan posted the company’s first campaign images, its Instagram followers jumped from 50 people to more than 20,000 in just a few weeks. “People called asking, ‘Can we have the contact for the press department?,” says Hassan. “But at the time, it was just me.” To easily handle media requests, she set up a [email protected] email address through the G Suite Admin console. And because she had used Gmail for several months before launching the company, using the business version of Gmail came naturally.

“Having that nubianskin.com address gave the business an air of professionalism, which we wouldn’t have had otherwise,” Hassan says. “Usually you need a whole IT department to set up new emails and passwords, but the Admin console removes the hassle of that process.”

Because Nubian Skin employees aren’t always in the showroom together, they need to be able to access business documents—like model photos and customer invoices—from anywhere. To help, Hassan securely stores business documents in Google Drive (which can be accessed on mobile, too) and works collaboratively with her finance team to process orders using Google Sheets. “Everyone has access to the spreadsheet so we can fill in what we need at any given point,” says Hassan. “This way, nothing slips through the cracks.” And when she needs to connect with colleagues quickly or face-to-face, she uses Hangouts Chat or Hangouts Meet. Chat makes it so employees are only a message away, and Meet helps her connect with people as if they’re in person at the office.

Five years after launching Nubian Skin, Hassan is a recognized entrepreneur who’s influenced other retailers to create more inclusive designs. “Even if you’re small, you want people to see you as a professional business,” Hassan says. “If you’re still growing, you need the building blocks to get you where you want to be. I can’t imagine working without G Suite.”

Use G Suite to make documents (and other tools) more accessible to people with disabilities

What does “accessibility” mean to businesses? It’s not just about making sure that your office or retail space is accessible to workers and customers who have mobility challenges, it’s also about making sure that your apps, digital tools and content are also accessible to everyone. 

Our product teams work everyday to ensure that tools like G Suite have built-in accessibility features. As a Program Manager, I help create those features and use many of them myself, as I’m blind. With October being Disability Awareness Month, this is a good opportunity to give an overview of product accessibility features that have been available for some time, as well as some new ones. 

Listen to content with help from screen readers. 
A screen reader is a helpful tool for people who are blind or have low vision. It provides methods to interact and control applications and also converts content on screen into spoken text. With a screen reader and keyboard shortcuts, you can read, edit, and comment on files. Getting started is easy: In Google Docs, Slides, or Sheets, go to the Tools menu, and in “Accessibility settings,” check the “Turn on screen reader support” box. Learn how toturn on screen reader support in Docs.

accessibility setting.png

G Suite supports several screen readers, including ChromeVox, Google’s own screen reader. ChromeVox is automatically built into Chrome devices, like Chromebooks. You just have to enable it. There are many other screen readers you can use with G Suite too; this support guide explains which screen readers work best with which web browsers.

Another way to listen to content is with Select-to-speak, a feature that reads content aloud to you using text-to speech (TTS) to verbalize highlighted content. To try it out, highlight text and activate the select-to-speak button. An example of a select-to-speak tool is the one built into ChromeOS.

See a summary of changes to your documents.
To make it easier for people to keep track of collaborators’ contributions in Google Docs (without having to actively watch every single change), we recently added a feature called Live Edits. Helpful if you use a screen reader or magnification, Live Edits periodically summarizes changes made to a document by collaborators within the sidebar. Learn how to comment and collaborate in Docs using a screen reader.

And for people who are sensitive to visual crowding, we also introduced eight different Lexend font families that have varied widths and spacing to accommodate different reading speeds. Read more details about these font options.

Use voice typing to “write” by speaking. 
Voice typing is a powerful productivity tool for everyone, including those who do not use a keyboard. It’s especially useful for people with low vision or with motor disabilities that may prevent them from using keyboards, mice or trackpads. When you speak into the computer’s microphone, voice typing uses artificial intelligence (AI) to convert your voice into text. In Docs, you can even use voice commands to select and edit text—for example, bolding words or cutting and pasting text.

727-GS-VoiceTypingDoc-DF.gif

Add closed captioning to presentation slides and meetings.
If you’re presenting slides or in a video meeting, you can add closed captioning. You don’t need to write the captions yourself—your computer’s microphone and Google’s machine learning tools automatically create captions as you speak and display them at the bottom of the screen. 

To use captions in Slides, click Present at the top right of your Slides screen, then click “CC” in the toolbar at the bottom of the screen. Learn how to present Slides with closed captioning.

An example of closed captions in Google Slides

This year, we also added support for automatic captions in Hangouts Meet so that when you’re in a video meeting, it’s possible to add captions, too. Keep in mind that the steps for adding captions to a video meeting can vary depending on the device you use–find instructions for closed captioning in video meetings for computers, Chrome devices, and Hangouts Meet hardware. 

Get through your inbox faster with accessible features.
Combing through emails can be tedious and time consuming. Previously, people using a screen reader would hear sender, date, subject, a snippet and more when arrowing thru their inbox. Now they have an alternative to reduce the verbosity of what they hear when they go through emails. By first arrowing right to a column such as “sender” or “subject,” people using screen readers can then down arrow and hear only that type of information. This helps people focus on only the information they need, saving time and reducing fatigue. 

We also recently introduceda new spelling and grammar tool in Gmail that’s powered by machine learning. It adds more functionality and a new keyboard interaction model (i.e. Left click and tab) while maintaining concise informative verbalizations and the existing right click keyboard interaction model.

The Blind Institute of Technology customizes G Suite apps for greater accessibility.
The need for accessibility can drive innovation. The Blind Institute of Technology (BIT), a nonprofit that helps businesses find accessibility solutions and hire people with disabilities, is staffed by people who are blind or have low vision. To help workers gather information from clients and job applicants, BIT staffers combined a screen reader, voice typing, and some under-the-hood programming in Apps Script. Workers can use voice typing to capture information, which is then collected in Forms and fed directly into Salesforce, where BIT stores customer records. Watch BIT’s presentation at Google’s recent Cloud Next conference for a deep dive into the custom Forms tools.

Making work more inclusive with G Suite
There are many ways to make documents, spreadsheets, and presentations accessible to more people—for example, adding alternative text to photos and graphics to help people who are blind or have low vision understand the purpose of your images. Check out tips on how to make your work more accessible

To learn more about these accessibility features and others, read this G Suite user guide to accessibility (or this G Suite Admin guide to accessibility if you’re an IT administrator) for a list of features that are built into G Suite tools. Also, watch this video about G Suite and Chrome Accessibility Features that I presented during Cloud Next ‘19.

New report analyzes the future of workplace productivity

TL;DR: we examined the future of work in a recent report. Download and read the findings

Look at the contemporary business landscape, and it seems like everything has changed in just a short amount of time. 

Today’s mid-career professional may have been in high school when the World Wide Web made the Internet a big commercial proposition. She likely started her career just before the dotcom bust, and, for nearly two decades, has witnessed the advent of big data, mobile, artificial intelligence, cloud computing, robotics, ecommerce, social media and more. Alongside the advent of these shifts in tech, the “office” has also transformed. From closed doors to cubicles to open plan, from typewriters to email to instant messaging, each transformation occurred in search of better information sharing and problem solving. 

Yet while it’s true that the world has changed, our ambitions as workers have not. The same things we’ve always wanted to get out of work remain: 

  • To be able to work fast, with fewer mind-numbing hassles in our day.
  • To be able to work smart, with quick access to the best possible information and the sharpest expertise.
  • To be able to chase the best ideas, and get our work recognized and improved for maximum impact.

While technology has increased the number of people we can connect with and how readily we can access new information, these opportunities can at times look like new challenges, especially if you rely on dated tools in the workplace. Nearly four in 10 U.S.-based business and IT leaders say their current systems make it harder, not easier, for their employees to work quickly. It’s like being asked to make carbon paper copies, when the rest of the world was first on email. 

Google’s latest report on the future of work examines challenges such as this, and how businesses can change their tools, workflows, and cultures to improve productivity and encourage innovation in the modern workplace. 

One of the interesting things about Google is that it was one of the first great companies to grow up assuming the internet as part of life. Consequently, this paved the way for the arrival of web-based email systems like Gmail, and productivity software to drive location-agnostic collaboration, like Google Drive or Docs. If you look at how these tools now incorporate advanced security and artificial intelligence for faster task execution, you’ll see a deep reflection of how work—and the world—has changed. People use these tools, however, because they meet human needs that have not changed.

Click here to download Google’s full report on the future of work, collaboration and productivity.

Celebrating 10 years of Apps Script: looking back on how it started

10 years ago this week we introduced Google Apps Script, G Suite’s scripting platform. What started as a small experiment has grown into one of Google’s most popular developer products. To commemorate the moment, I sat down with Mike Harm, the creator of Apps Script, to reflect back on how it all got started. 

Mike, what inspired you to build Apps Script?
I taught an “Intro to Computers” class when I was a graduate student at Northeastern University. The course was intended for students outside the computer science department, so I ended up teaching Pharmacy majors. The course covered everything from drag-and-drop to office productivity tools, building up to a final project using Hypercard, where the assignment was just “make something cool.” Those students produced so many interesting things: an interactive recipe book, a Led Zeppelin trivia game, an animated, choose-your-own-adventure story. It made a lasting impression on me. Once you strip away all of the technical hoops, people can build amazing things, even if they’re not programmers.

The same revelation happened later in my career. This time, in the workplace. I was working in the IT department of a New York hospital that had built a custom application server running server-side JavaScript, years before Node.js was created. This engine powered a lot of their approvals systems. Because of how accessible JavaScript was, business analysts could write the code themselves instead of waiting on IT to implement it. 

How did you use this experience to create Apps Script at Google?
I worked on the Google Sheets team when I first joined Google in 2007. At the time, the team was evaluating how to make our spreadsheet APIs easier to use. With Hypercard and JavaScript still bounding around in my head, I kicked off a stealth, side project to build a rough demo of what a JavaScript-based scripting language for spreadsheets could look like. This was the start of Apps Script.

Why did you choose JavaScript as the programming language?
I was certainly influenced by my previous experience with JavaScript, but also saw how popular the language was becoming in web programming. Specifically, there were a lot of non-traditional programmers using JavaScript to add small bits of dynamic functionality to websites, and the syntax was becoming well known. I still had Hypercard on my mind, and I wanted a language that anyone could code in.

I wanted a language that anyone could code in.

Were there challenges building a tool for non-traditional programmers?
We had many passionate discussions about whether to adopt traditional coding paradigms, like versioning, or to build something more accessible for non-programmers. Even I had trouble keeping my eye on the pharmacy students I originally intended to serve.

We hosted hackathons to help us keep in tune with what was really needed. I remember an 11-year-old demoed a simple celsius to fahrenheit custom function. I totally swooned. From that point forward, when we discussed how to shape Apps Script, I’d think about that kid—for some, a red error message is an occupational hazard, for him, it was personal.

Can you share other examples of how you use Apps Script?
I turn to Apps Script whenever I say to myself, “I wish [this app] could do that!” For example, I built a web app that lets me see all of my meetings for the week, check a box next to the ones to skip, and then click one button to cancel them—works great when I travel to other offices.

Companies can be hesitant to hand over low-code app development broadly to their employees. Do you think they should embrace this democratization?
I’ve thought about this a lot. I feel that they should embrace it, but that there should be guard rails. A good metaphor is that we should allow people to plug in a teapot at their desk, but not an arc welder that could black out the building. When dialing in our quotas for Apps Script, we made sure that users had the amount of power they needed to get important work done, but not so much power that it would become a nightmare for IT teams to manage. 

10 years in the making, Apps Script has inspired people to customize G Suite apps in many different ways—check out our video library to see it in action (along with sample code) and start building today.

G Suite collaboration, for Dropbox users

By now, it is clear that the cloud is not binary—choices aren’t limited to one cloud or another, or to one service or another. Businesses require solutions that converge so that their workers can collaborate more effectively.

This is why we’ve built an open ecosystem in Google Cloud. We value and encourage partnerships within this ecosystem because these joint innovations are what’s most helpful to our customers. Today, we are delighted to announce that Dropbox customers can access the collaborative power of G Suite—specifically in Google Docs, Sheets and Slides.

Dropbox for G Suite creates a unified experience for the millions of customers who are working jointly in both services by integrating core G Suite products—Google Docs, Sheets, and Slides—with Dropbox’s content and collaboration platform. Together, employees can:

  • Create and store Docs, Sheets and Slides in Dropbox alongside other traditional files.
  • Use Google Docs, Sheets and Slides to edit Microsoft Office file types stored in Dropbox, without having to change file formats.
  • Add Google Docs, Sheets, and Slides to shared Dropbox folders which will then  automatically inherit the same sharing permissions.
Create, organize, and share Google Docs, Sheets and Slides from Dropbox to boost productivity, reduce time spent switching between tools, and keep your content in one centralized place.

Improving collaboration at Dow Jones
Dow Jones, a publishing and financial information company that has operated for more than  200 years, uses G Suite as its standardized enterprise productivity solution and is an early adopter of Dropbox for G Suite.

“Dropbox for G Suite has given our employees the flexibility to use the tools they love, while creating a simple, connected way for them to share and work together,” says Shaown Nandi, CIO of Dow Jones and Head of North America Shared Services for News Corp. “This is possible in large part because of the interoperability that Google has added with its partners like Dropbox, so we’re excited to see these integrations expanding.”

This integration compliments joint services that already exist between Dropbox and G Suite, including ways to access files saved in Dropbox directly from Gmail ways to share files to Gmail from Dropbox, and ways to provision and deprovision Dropbox users directly from the Cloud Identity or G Suite Admin console to improve employee on-boarding workflows and user lifecycle management. Learn more about Dropbox for G Suite.

Building an open cloud
G Suite has hundreds of integrations to help businesses work more effectively, and we’re continuing to invest in both the G Suite platform and its ecosystem of partners. When it comes to content collaboration, G Suite already integrates with numerous partners to help enterprises organize and use content more effectively, including Dropbox, DocuSign, Salesforce, Evernote, LucidChart and more.

Outside of content collaboration, G Suite has hundreds of integrations to help businesses work more effectively. Learn more on our website.

Mail merge with the Google Docs API

Posted by Wesley Chun, Developer Advocate, Google Cloud

Students and working professionals use Google Docs every day to help enhance their productivity and collaboration. The ability to easily share a document and simultaneously edit it together are some of our users’ favorite product features. However, many small businesses, corporations, and educational institutions often find themselves needing to automatically generate a wide variety of documents, ranging from form letters to customer invoices, legal paperwork, news feeds, data processing error logs, and internally-generated documents for the corporate CMS (content management system).

Mail merge is the process of taking a master template document along with a data source and “merging” them together. This process makes multiple copies of the master template file and customizes each copy with corresponding data of distinct records from the source. These copies can then be “mailed,” whether by postal service or electronically. Using mail merge to produce these copies at volume without human labor has long been a killer app since word processors and databases were invented, and now, you can do it in the cloud with G Suite APIs!

While the Document Service in Google Apps Script has enabled the creation of Google Docs scripts and Docs Add-ons like GFormit (for Google Forms automation), use of Document Service requires developers to operate within the Apps Script ecosystem, possibly a non-starter for more custom development environments. Programmatic access to Google Docs via an HTTP-based REST API wasn’t possible until the launch of the Google Docs API earlier this year. This release has now made building custom mail merge applications easier than ever!

Today’s technical overview video walks developers through the concept and flow of mail merge operations using the Docs, Sheets, Drive, and Gmail APIs. Armed with this knowledge, developers can dig deeper and access a fully-working sample application (Python), or just skip it and go straight to its open source repo. We invite you to check out the Docs API documentation as well as the API overview page for more information including Quickstart samples in a variety of languages. We hope these resources enable you to develop your own custom mail merge solution in no time!

Everyday AI: beyond spell check, how Google Docs is smart enough to correct grammar

Written communication is at the heart of what drives businesses. Proposals, presentations, emails to colleagues—this all keeps work moving forward. This is why we’ve built features into G Suite to help you communicate effectively, like Smart Compose and Smart Reply, which use machine learning smarts to help you draft and respond to messages quickly. More recently, we’ve introduced machine translation techniques into Google Docs to flag grammatical errors within your documents as you draft them.

If you’ve ever questioned whether to use “a” versus “an” in a sentence, or if you’re using the correct verb tense or preposition, you’re not alone. Grammar is nuanced and tricky, which makes it a great problem to solve with the help of artificial intelligence. Here’s a look at how we built grammar suggestions in Docs.

The gray areas of grammar

Although we generally think of grammar as a set of rules, these rules are often complex and subjective. In spelling, you can reference a resource that tells you whether a word exists or how it’s spelled: dictionaries (Remember those?).

Grammar is different. It’s a harder problem to tackle because its rules aren’t fixed. It varies based on language and context, and may change over time, too. To make things more complicated, there are many different style books—whether it be MLA, AP or some other style—which makes consistency a challenge.

Given these nuances, even the experts don’t always agree on what’s correct. For our grammar suggestions, we worked with professional linguists to proofread sample sentences to get a sense of the true subjectivity of grammar. During that process, we found that linguists disagreed on grammar about 25 percent of the time. This raised the obvious question: how do we automate something that doesn’t run on definitive rules?

Where machine translation makes a mark

Much like having someone red-line your document with suggestions on how to replace “incorrect” grammar with “correct” grammar, we can use machine translation technology to help automate that process. At a basic level, machine translation performs substitution and reorders words from a source language to a target language, for example, substituting a “source” word in English (“hello!”) for a “target” word in Spanish (¡hola!). Machine translation techniques have been developed and refined over the last two decades throughout the industry, in academia and at Google, and have even helped power Google Translate.

Along similar lines, we use machine translation techniques to flag “incorrect” grammar within Docs using blue underlines, but instead of translating from one language to another like with Google Translate, we treat text with incorrect grammar as the “source” language and correct grammar as the “target.”

gsuite_grammar.jpg

Working with the experts

Before we could train models, we needed to define “correct” and “incorrect” grammar. What better way to do so than to consult the experts? Our engineers worked with a collection of computational and analytical linguists, with specialties ranging from sociology to machine learning. This group supports a host of linguistic projects at Google and helps bridge the gap between how humans and machines process language (and not just in English—they support over 40 languages and counting).

For several months, these linguists reviewed thousands of grammar samples to help us refine machine translation models, from classic cases like “there” versus “their” versus “they’re,” to more complex rules involving prepositions and verb tenses. Each sample received close attention—three linguists reviewed each case to identify common patterns and make corrections. The third linguist served as the “tie breaker” in case of disagreement (which happened a quarter of the time).

gsuite_grammar_suggestion.jpg

Once we identified the samples, we then fed them into statistical learning algorithms—along with “correct” text gathered from high-quality web sources (billions of words!)—to help us predict outcomes using stats like the frequency at which we’ve seen a specific correction occur. This process helped us build a basic spelling and grammar correction model.

We iterated over these models by rolling them out to a small portion of people who use Docs, and then refined them based on user feedback and interactions. For example, in earlier models of grammar suggestions, we received feedback that suggestions for verb tenses and the correct singular or plural form of a noun or verb were inaccurate. We’ve since adjusted the model to solve for these specific issues, resulting in more precise suggestions. Although it’s impossible to catch 100 percent of issues, we’re constantly evaluating our models at Google to ensure bias does not surface in results such as these.

Better grammar. No ifs, ands or buts.

So if you’ve ever asked yourself “how does it know what to suggest when I write in Google Docs,” these grammar suggestion models are the answer.  They’re working in the background to analyze your sentence structure, and the semantics of your sentence, to help you find mistakes or inconsistencies. With the help of machine translation, here are some mistakes that Docs can help you catch:

gsuite_grammar_suggestion_models_1.jpg
gsuite_grammar_suggestion_models_2.jpg
gsuite_grammar_suggestion_models_3.jpg

Evolving grammar suggestions, just like language

When it comes to grammar, we’re constantly improving the quality of each suggestion to make corrections as useful and relevant as possible. With our AI-first approach, G Suite is in the best position to help you communicate smarter and faster, without sweating the small stuff. Learn more.

Why so tense? Let grammar suggestions in Google Docs help you write even better

If you’re working against deadlines to create documents daily (how’s that for alliteration?), having correct grammar probably isn’t the first thing on your mind. And when it is, it seems there’s almost always a contested debate about what is correct (or “which?”). Even professional linguists have a hard time agreeing on grammatical suggestions—our own research found that one in four times linguists disagree on whether a suggestion is correct.

We first introduced spell check in Google Docs to help folks catch errors seven years ago, and have since improved these features so that you can present your best work. Today we’re taking that a step further by using machine translation techniques to help you catch tricky grammatical errors, too, with grammar suggestions in Docs (which we first introduced at Google Cloud Next last year).

G Suite Basic, Business, and Enterprise customers will start to see inline, contextual grammar suggestions in their documents as they type, just like spellcheck. If you’ve made a grammar mistake, a squiggly blue line will appear under the phrase as you write it. You can choose to accept the suggestion by right-clicking it.

Grammar Suggestions with Device GIF

“Affect” versus “effect,” “there” versus “their,” or even more complicated rules like how to use prepositions correctly or the right verb tense, are examples of errors that grammar suggestions can help you catch. Because this technology is built right into in Docs, you don’t have to rely on third-party apps to do the work.

How it works

When it comes to spelling, you can typically look up whether a word exists in the dictionary. Grammar is different. It’s a more complex set of rules that can vary based on the language, region, style and more. Because it’s subjective, it can be a harder problem to tackle using a fixed set of rules. To solve the problem, we use machine translation to build a model that can incorporate the complexity and nuances of grammar correction.

Using machine translation, we are able to recognize errors and suggest corrections as work is getting done. We worked closely with linguists to decipher the rules for the machine translation model and used this as the foundation of automatic suggestions in your Docs, all powered by AI.

In doing so, machine translation techniques can catch a range of different corrections, from simple grammatical rules such as how to use “a” versus “an” in a sentence, to more complex grammatical concepts such as how to use subordinate clauses correctly.

Using artificial intelligence to make work easier

Google’s machine intelligence helps individuals collaborate more efficiently everyday in G Suite. If you’ve ever assigned action items or used the Explore feature to search for relevant content to add to your Docs, you’ve tried the power of AI first hand.

Happy writing!

Why so tense? Let grammar suggestions in Google Docs help you write even better

If you’re working against deadlines to create documents daily (how’s that for alliteration?), having correct grammar probably isn’t the first thing on your mind. And when it is, it seems there’s almost always a contested debate about what is correct (or “which?”). Even professional linguists have a hard time agreeing on grammatical suggestions—our own research found that one in four times linguists disagree on whether a suggestion is correct.

We first introduced spelling and grammar check in Google Docs to help folks catch errors seven years ago, and have since improved these features so that you can present your best work. Today we’re taking that a step further by using machine translation techniques to help you catch tricky grammatical errors, too, with grammar suggestions in Docs (which we first introduced at Google Cloud Next last year).

G Suite Basic, Business, and Enterprise customers will start to see inline, contextual grammar suggestions in their documents as they type, just like spellcheck. If you’ve made a grammar mistake, a squiggly blue line will appear under the phrase as you write it. You can choose to accept the suggestion by right-clicking it.

Grammar Suggestions with Device GIF

“Affect” versus “effect,” “there” versus “their,” or even more complicated rules like how to use prepositions correctly or the right verb tense, are examples of errors that grammar suggestions can help you catch. Because this technology is built right into in Docs, you don’t have to rely on third-party apps to do the work.

How it works

When it comes to spelling, you can typically look up whether a word exists in the dictionary. Grammar is different. It’s a more complex set of rules that can vary based on the language, region, style and more. Because it’s subjective, it can be a harder problem to tackle using a fixed set of rules. To solve the problem, we use machine translation to build a model that can incorporate the complexity and nuances of grammar correction.

Using machine translation, we are able to recognize errors and suggest corrections as work is getting done. We worked closely with linguists to decipher the rules for the machine translation model and used this as the foundation of automatic suggestions in your Docs, all powered by AI.

In doing so, machine translation techniques can catch a range of different corrections, from simple grammatical rules such as how to use “a” versus “an” in a sentence, to more complex grammatical concepts such as how to use subordinate clauses correctly.

Using artificial intelligence to make work easier

Google’s machine intelligence helps individuals collaborate more efficiently everyday in G Suite. If you’ve ever assigned action items or used the Explore feature to search for relevant content to add to your Docs, you’ve tried the power of AI first hand.

Happy writing!