It’s interesting watching how automated assistants are evolving at search engines. A recent study that is worth looking at is this one from Perficient Digital: Rating the Smarts of The Digital Personal Assistants of 2019
The automated assistant that I am paying the most attention to is the one from Google, because I have an android phone, and a Google Smark Speaker and they often work together to provide search results to me.
Google was granted a patent recently about automated assistants, which you may also know as “personal assistants”, “mobile assistants”, or “chatbots.”
The patent looks at what it calls “resource and/or interface constraints” that may cause an automated assistant to have drawbacks when providing search results tor present to a searcher. The patent is about how those limitations might be overcome when showing search results.
Problems with Personal Assistant Search Results Being Presented to Searchers
An automated assistant implemented via a “chat” type graphical user interface providing a lot of results may:
- Make such an interface cluttered
- Make dialog harder to follow
- May consume a relatively large amount of computational resources.
For instance, we are told that computational resources may be consumed because of the assistant simultaneously rendering a large number of results and/or “scrolling” and/or other actions that may be required to view a large number of results.
And the making of the search results viewable in an interface and/or application that is separate from the automated assistant interface and/or application may also consume greater user and/or computational resources.
Switching to another interface or application may distract searchers from ongoing dialog with the automated assistant.
An Automated Assistant may also be restricted to an audible user interface output and accepting spoken user interface input. When that happens, directed navigation of search results by a searcher may not be possible or may be limited.
The focus of the automated assistant search results patent is on improving search results from automated assistants for searchers.
One approach may enable a user to provide textual input to navigate the search results within the spoken dialog and handle the interface constraints associated with using the automated assistant.
The patent tells us that there are some advantages from following the process described in the patent and that those are:
- Enable fewer search results to be provided to the user, while still satisfying the informational needs of the user
- Navigating the search results within the dialog may enable a user to freely navigate (optionally non-sequentially) forward and/or backward through search results, without necessitating that all of the search results be presented simultaneously and/or that the entirety of one or more of the search results be presented
Search Results Formatting for an Automated Assistant
The patent tells us about an example, that a searcher may cause textual input to be provided to an automated assistant during a dialog between the user and the automated assistant. That text may initiate the dialog or maybe a continuation of a previously initiated dialog. The textual input may be in natural language free-form input, such as textual input that is based on user interface input generated by the user either typed in or spoken.
This is when Google asks me if I would like to have links to news items that it has read headlines from, so that I can follow those and read the whole articles, in addition to the summary that the Google automated assistant provided to me. It offers to send those links to me on my phone, which my smart speaker knows is (often) right next to it
The patent does provide more details that fit in with what I added to the example above:
For example, in response to textual input of “news headlines”, a database of news story documents may be searched, and a plurality of search results obtained that are each based on a respective one of a plurality of recent news story documents identified in response to the search. The obtained search results may be selected and/or ranked based on various signals, such as the popularity of the search results, a degree of matching between the search parameters and the search results, attributes of the user, etc.
The patent provides several alternative approaches to showing search results, and interacting with a searcher in the display of those, and providing alternatives gives it a chance to evolve in how it shows results to searchers.
- Audible presentation of search results
- Showing Search Results on a Display
- Showing Search results after additional dialog between a searcher and an automated assistant, such as “should I send those results to your phone?”
This patent can be found at:
Using user input to adapt search results provided for presentation to the user
Inventors: David Kogan and Bryan Horling
Assignee: GOOGLE LLC
US Patent: 10,481,861
Granted: November 19, 2019
Filed: August 30, 2016
Methods, apparatus, and computer-readable media related to the interaction between a user and an automated assistant during a dialog between the user and the automated assistant. Some implementations are directed to adapting a graphical and/or audible presentation of search results provided by the automated assistant for presentation to the user. The adaptation may be in response to attribute(s), of one or more of the search results, referenced in spoken and/or typed textual input provided by the user during the dialog. Some of those implementations may enable a user to provide textual input to navigate the search results within the dialog and within the resource and/or interface constraints associated with the dialog. Some of those implementations may additionally and/or enable adapting, based on textual input provided by a user to the automated assistant, when and/or whether search results having certain attributes are provided to the user by the automated assistant.
Automated Assistant Search Results and More Takeaways
There are several things that the Google Automated Assistant can do.
Finding ways to show search results is the focus of this patent, but it is also worth looking at some of the other features involving the automated assistant.
Google provides a queryless feed showing results in response to a searcher’s interests in a feature that is being called Google Discover. You can optimize for Google Discover as described in Google Discover Optimize your content for Discover. They have also added a performance report for Discover to Google Search Console.
One place that we see Google assisting site owners is when they released Speakable Schema as a beta product for News sites. So if you own a news site, and you have some content that should ideally be spoken in response to a query, marking that content up with Speakable Schema would be ideal. The Google Developer’s page doesn’t say that this feature will become available for more than just news sites, but it might someday, so keep your eyes open for when that may happen.
Spoken Search Results:
Google has been researching Spoken results, and published this post on one of their blogs in March of last year: Expressive Speech Synthesis with Tacotron. It tells us about some of the efforts Google has been making with text to speech efforts and adding pauses and emphasis (prosody) in that speech.
Google Duplex Calls:
People who find your business using Google Search, Maps, or the Google Assistant can ask it to call your business, as described on this page, to do things like book an appointment: About phone calls from the Google Assistant. It is possible to opt out of receiving these automated calls to your business.
Google Assistant Actions:
You can create actions that a Google Assistant may be able to perform like turning on lights or playing a trivia game. More details about that can be found at: Actions SDK basics
Google Assistant News:
A feature that was launched a week ago has Google taking advantage of Digital audio, by letting the automated assistant provide news, as they describe in Hey Google, play me the news
Added December 13, 2019, I recently wrote about a related patent involving the Google Assistant, in the post The Google Assistant and Context-Based Natural Language Processing on how the dialog system behind the Google Assitant works and understand the contexts of user requests.