When you live with something for a while, it can be difficult to detect little changes that occur over time. The subtle differences and small tweaks can slip by without notice. I am definitely having this experience with Siri in my daily use over the last few months. I can’t tell you when both Siri and the on-board Voice Dictation started replacing and correcting words after the fact based on grammar rules, but I see that it’s happening on a regular basis now. One direct example I noticed recently would be Dictation replacing “there” with “their” based on the context of the sentence after I had dictated a few more words. I’ve seen the same with “your” and “you’re,” as well. This makes voice dictation far more flexible and usable, since I don’t have to go back and edit as often.
I have also noticed that both Siri and Voice Dictation are doing better at recognizing the spellings of the names of my contacts and typical locations when I add them to Calendar entries and Tasks. Both are also doing a much better job of surfacing the apps I am interested in when I do a Spotlight Search, based on time of day and my location. This wasn’t the case when these “Siri Intelligence” features appeared last year, but I now find myself using this method to get to a handful of apps that are located in folders on different Home Screens without having to do anything more than a swipe down and a single tap. These are the kinds of small changes that we should expect to see over time as machine learning works its magic in the background.
These bits of progress don’t cover over the fact that Siri isn’t as flexible or adaptable as other digital assistants, and isn’t capable of finding answers to even some common questions beyond searching Google. When you consider that Apple was in this space before anyone else and now finds itself behind, the progress above isn’t even worth a pat on the back. Apple is playing catch-up, and will be for a while yet. That said, I found reports of their renewed commitment to Siri, and the AI and Machine Learning backbone behind it from late last year encouraging. They are finally putting the emphasis and investment into Siri and machine learning that is befitting a high priority feature of iOS.
While it’s hard to say how long it will take for these investments to pay off in a big way, it looks like users may finally be seeing some fruits of these labors. A recent survey on satisfaction with digital assistants conducted by CBT Nuggets shows evidence that Siri is turning a corner in some key customer satisfaction metrics. Siri actually bested Google Assistant, Cortana, and Alexa in terms of overall satisfaction, as well as in the more specific categories dealing with use in quiet and noisy areas. Surprised? I certainly was, and it’s why I’m writing about it. Be sure to follow the link and take a look at how each of the major digital assistants scored and raked in several different categories that they polled users about. It definitely is an interesting read if you are interested in digital assistants and how they stack up.
Photos Sourced from CBT Nuggets
As you can see if you look at the infographics in the article, Siri came out on top the most across the questionnaire. However, it wasn’t all good news for the original digital assistant. Apple actually struggled quite a bit in one key area where most of us would expect them to be strong- Music. In fact, this is completely at odds with the claims that were made during the HomePod reveal at WWDC. We were told that Siri would become an expert “musicologist”, and that it would be the primary interface for requesting music on the coming device.
Now, it could be that Apple has not rolled out the feature set for Siri that they demoed on stage back in June. Maybe all of that is still forthcoming, and will work much better than what we have now when turned lose in the cloud. However, it’s also possible that Siri’s enhanced music features weren’t ready for prime time and that’s why they pushed the HomePod’s launch back. Based on these survey results, it wouldn’t surprise me if it was at least a reason. The most important thing is that it works at launch. If it does, then all is forgiven. However, it is worrisome to see Apple struggling in such a key area with a soon-to-be-released product that will absolutely depend on Siri’s rapid improvement.
As for the results in this survey, while they are detailed, they are not scientific. This is just a simple survey with answers broken down in different way to help readers gain some insights. Hopefully, this or another group will do a larger digital assistant survey in the future that will give us an even deeper look into how people think of them and use them. I won’t go as far as saying we shouldn’t trust these results, but this is the kind of study that warrants further study, rather than one that becomes a definitely answer on the state of digital assistants.
So what do you think? How does your experience with Siri match up with this survey and its results, specifically its results on Siri? I am very interested to know, because most of those who have commented about Siri on other articles here have been critical for various reasons. Pro or con, I would love to hear your take. Let me know in the Comments section below, on Flipboard, on our Facebook page, or on Twitter @iPadInsightBlog.