Changes we're making to Google Assistant
blog.google
blog.google
I've already decided that I'm done with Google's smart assistant stuff in any case. I have a Google Home with a screen in my kitchen and the most-used feature (aside from just existing as a photo frame) was an integration with a really useful shopping list app called AnyList. It certainly wasn't complex, we'd say "Hey Google, add <x> to the shopping list" and it would do it. But it was very useful: I'd have something in my hands I just pulled from the fridge (e.g. milk) and be able to add it to the list without interrupting what I'm doing. If it had to wait until I was done for me to pull out my phone I'd inevitably forget.
Then one day Google decided to disable that integration. Now the only shopping list you can add to is one Google provides (which naturally has way fewer features than AnyList). They've never provided even the remotest defense for why they've removed it, it's very obviously to lock us into the Google ecosystem. So our Google Home is now a glorified photo frame that plays music from time to time (and even then prioritizes cover versions and YouTube videos over actual songs, presumably because $$$)
Never thought amazon would have the better assistant ecosystem. Google already has my email and calendar and other stuff, Amazon I have to auth into them like a third party. But it's become true
There are some gpt and llama experiments you can run there, but it's finally getting to the point where a local assistant you host yourself might be viable soon
Getting audio into it and having a little speaker so it can talk back is both the easiest and the hardest part. I need to play around more with some esp32s and microphone modules
Boom, done. Synced with my wife too. This is using the Apple Reminders though, as it seems to have less friction than third party apps…
There is no command ontology anymore. Consequently, discovering commands in-app is UX anathema.
"Use the internet, buddy. One of our customers probably documented how-to somewhere on it."
I recently had to contact windows help, and they were useless. The windows forum points to a third party software that I am convinced is malware.
And while I'm ranting, why can't the download folders in windows be set to group by "none" and sort by "name" by default? Why are both "most recently edited" by default? AND WHY CAN'T I CHANGE IT MICROSOFT?
[0]: https://github.com/LesFerch/WinSetView
[1]: https://github.com/LesFerch/WinSetView/blob/main/README-more...
Device > Settings > Sounds > Notification > none
Everything in my post was Alexa-specific. Someone who has the stated problem can follow that precise and non-obvious menu tree within the Alexa app, and it will solve their problem. It's a device-specific setting, so can be done on the kitchen Alexa without affecting other units.
Both "notification" [1] and "Do Not Disturb" [2] in my post were Alexa-specific terms that are unrelated to their mobile phone brethren.
[1] https://storables.com/home-appliances/home-automation-appliances/how-to-turn-off-notification-on-alexa/
[2] https://www.lifewire.com/how-to-use-alexa-do-not-disturb-mode-4788343> okay Google set a timer for 15 mins
Certainly, and did you know I can also set an alarm for 6 am in the kitchen?
> Fuck off Google
Setting a timer and have it remind you that you could have set an alarm instead is annoying, but there have been multiple examples where we set a timer and the Google home mini seems to forget it.
> okay Google, set a timer for 1 hour
>> Certainly, and did you know that I can also look up Wikipedia entries for you.
>> Timer for 1 hour has been set and is starting now
> okay Google, how much longer left on timer
>> You don't appear to have any timers set at this time.
I'm pretty sure that our robot overlords will either forget, get distracted with generating mis-information, or become artists painting portraits of furries.
EDIT: Yep its still there! https://www.amazon.com/gp/help/customer/display.html?nodeId=...
Alexa recently started responding with "Good afternoon! <the normal response>" and it irks me more than it probably should. I've looked to see if I can turn it off and can't find the option.
I don't need pleasantries from a machine.
notice how young kids associate ok google with a way to install games and reach their YouTube videos.
all those feature are for that end. you're not the target market. Good night!
If you say "Alexa, stop by the way", it'll get the device to stop responding with follow-ups for ~24h. I ended up creating a routine that runs every night at 4am to lower the volume to zero, automatically say "stop by the way" to the device, and then raise the volume a minute later, and Alexa has stopped with the follow ups
In other words, they have just closed down that entire ecosystem. I don't think there's an antitrust angle here: Forcing e.g. Apple to allow third-party apps or app stores seems within reach (at least in the EU); this would be more like forcing Apple to enable apps in a world where they don't even exist (anymore).
Here, "conversational" means your voice command should prompt the app to ask you a question, driving a state machine. (Set a reminder. When do you want to be reminded?)
I haven't tried to use it, but the docs indicate CRUD-over-voice ("built-in intents") is unaffected.
https://developers.google.com/assistant/ca-sunset#what_will_...
https://developer.android.com/reference/app-actions/built-in...
“What’s the weather @ time” is inadequate. I live in California(albeit not on the coast or in the valley).
The temperature in Nyagen, Russia is current -31 degrees...
I'm an American but I've been to Nijmegen so I think my pronunciation was ok.
Completely opens up these displays to be a virtually unrestricted canvas for doing or displaying anything you want.
That said, we are building a Home Assistant integration at Tidbyt as well.
https://github.com/pkulak/tidbyt
It ended up working great!
So the difference wrt discovery is that you only have to gesture at what you wanna do and, if a matching action exists, there is a chance it will be understood.
I'd wager we'll see a renaissance of voice assistants with LLMs, especially once the good-enough ones can run on device.
Ow, right in the AARP flyer.
Not really conversations though, more like accessible commands. I don’t think we get conversations until the tech improves and the latency goes way down, meaning on device processing of speech at the very least.
I wonder how well they’ve optimised Siri for a just few specific use cases like this, because nothing else seems to reliably work on it.
Until now I have only been saying simple commands like "countdown n minutes" or "call my favorite wife" (I only have one but thankfully she is also my favorite.)
I would be very sad if these were to disappear or to stop working as well as it does now (I like that if I say “remind me tomorrow” when it’s right past midnight it asks for confirmation that I actually mean the same day since it’s “already tomorrow”)
The device is out of support, it has its glitches (e.g. it's still able to play and pause podcasts, but after pausing it would say "Sorry. something's gone wrong", if asking it to snooze an alarm it would say "OK, alright" and then "snoozing for 10 minutes"), looking forward (not!) to Google making a breaking change and it being e-waste soon.
Hah, in the imaginary future where we have cyborg assistants instead of climate destruction, my robot butler will pour half a cup of tea, go back to the kitchen, and then return to pour the second half of the cup, and I'll tell my guest, "Yeah, it's a 5 year old model, the startup that made him went bust so I'm using a firmware from a Ukranian forum... Джевс, можна мені трохи цукру?"
I think for most folks, affordability is more about reality than mentality.
Personally, I'd have no use for a free iphone. Android handsets are the only option for the functionality I use. Been that way since WinMo6.
Apple has exposed Siri to us as developers in several ways, one of which is called “Siri intents.”
The long and short of it is that app developers can hook into the Siri system so users can use Siri commands to control certain parts of third-party apps.
This means that users aren’t locked into using, say, Apple’s own Reminders app, and they can tell Siri to add things to third-party to-do list apps.
Users also have the option to create their own workflow via Shortcuts and give it a command phrase that Siri then responds to.
I assumed something similar was available on Android, but the responses indicate that this might not be the case.
I think the broader context here is that Google is downsizing the current Assistant team in preparation for an LLM-based replacement, perhaps once Gemini has rolled out.
If this doesn’t seem obvious in 1yr I’d say the hypothesis is likely wrong.
I don't think voice is a good interface. "It chats like a human" is the lowest possible hanging fruit in terms of product design, and bets everything on the smarts of the tech that's behind it.
We are so used to tooling faster than voice. Keyboards and taps are very, very fast. I want digital assistants as smart AND as fast as that, not something smart but incredibly slow to interact with because it needs to dumb itself down to human speech I/O.
To me, this is also not about modality or making it more generic. I just don't want an anthropomorphized smart-ass assistant. I want smart tools that actually assist me directly, no chat.
People hate call centers. I don't want to have a conversation with some human or human-parity AI at my airline, I want to change my flight in 3 clicks. In retail and fast food, self-checkout and online or kiosk ordering is popular for a reason.
I've found myself feeling those same feelings when being forced to have conversations with a chatbot. Chatting with ChatGPT just for fun can be fun, but it can be just as painful as a call center if you need to get something specific done. In order to cancel a hotel reservation, I was having to chat with some bot. It made it into a whole conversation, with brief pauses. It should have been 3 clicks.
LLMs can plausibly solve both.
Also, I don't want to engineer a prompt to convince my assistant to do my bidding. That's just hand coding action trees with extra steps.
I have an Alexa list called “Grocery” that does a 2-way sync, and use it to accelerate my online Safeway orders via AnyList.
You can sync to the default Amazon Shopping list too, but I find it works better if you have a secondary one specifically for sync.
> Discoverability is zero
Today's voice assistants are the opposite. They are unreliable and completely unstable. They don't have a clear list of commands they understand, let alone some sort of menu system - the documented commands on the manufacturer's websites often don't work. They also randomly change what they can do: stuff stops working for no obviously reason.
For example - yesterday, my son's Nest Audio suddenly refused to set a music alarm, claiming "this device doesn't support that feature yet" (it's literally a speaker for music...). It worked the day before.
Identifying the problem is often the first step in fixing it.
I can not for the love of my life get Siri to resume play in my native tongue. It is supported by Siri, I just have no clue on earth on what phrases she accepts. Tried many, started feeling dumb. Turned off Siri.
Every single one of those things have been deprecated and removed for… reasons?
All we can do with the speaker now is ask it to play music, get the weather, or set a timer.
Surprising to hear that Google assistant is now equivalent to Siri!
If I recall correctly, "add to my list" originally went to shoppinglist.google.com. They rolled it into Keep[1] ("note to self," tags, sharing).
It works well. However...
In 2016, I figured "tell me showtimes for $MOVIE in $CITY" should work. It still doesn't work. It may never work, because "robot butler with ads" is no one's idea of the future. (Maybe Jeff.)
A fire-and-forget voice action like "add to list" should still work(?). For conversations[2], we can pray Gemini is less infuriating.
[1] https://play.google.com/store/apps/details?id=com.google.and... [2] https://developers.google.com/assistant/ca-sunset
Same for the weather prompt someone else posted above and contrary to their experience it showed me exactly what I expected.
I'm a bit confused if those examples are something that didn't work in the past or if it's something that somehow doesn't work properly in the US right now.
There's a Nest mini (2nd gen) two meters away, but the device that answers is the Google Home 2016 in a different room.
First attempt: "Okay, ten minutes. And that's starting... Now."
Second attempt: "Sorry, I don't understand."
Third attempt: bingo.
Further trials seem to work, until they're interrupted by the 10 minute timer.
My partner comes downstairs to ask why I'm yelling STOP.
I remember my exact words when I bought this thing:
I have seen the future, and it sucks.
Also a lot of people rent studios and one bedrooms or otherwise have open floor plans.
Not to mention I think the paranoia about these home assistant devices is overblown. Do you use a smartphone? In your intimate kitchen?? It has just as much capability to listen to you.
We have a saying that loosely translates to “If the barn burns down, so does the house”. The meaning is that it’s not really wise to let the house burn down if the barn is burning.
Non user-hostile technology would happily solve this problem: you pull from the fridge what is about to finish or before throwing the empty container into the trash, put it in front of a mini camera mounted on the fridge, image (possibly also barcode) recognition would identify the product, then you say a magic word and it would be added it to your list directly on the phone, then software would calculate the best path to the shops having in stock the products in the list. Sadly, this would be hijacked by businesses in no time by making it dependent from some cloud services then selling your data to other advertisers or by pestering you with offers of similar competing products. As of today, I still prefer a pen and a piece of paper over any type of automation.
If anybody's curious, repo with most of the code is here: https://github.com/csimpf/keep2anylist. Happy to answer any questions about it!
I want to use Microsoft ToDo for my tasks. Why? I use Outlook everywhere. I don't want have to keep track of tasks somewhere else.
I'm looking forward to the EU forcing companies to allow more integrations.
Assistants were SO much more usable when they would require strict trigger language, and interpret it literally and formulaically. Yes you would have to learn its language, but once you did, you could actually accomplish what you wanted to accomplish, unlike today.
Google has turned into these little micro clusters of ideologies and functional philosophies, leading to a horrifically fragmented mess of google branded products. All which come loaded now with end-of-life anxiety as google just farts out products that seem to dissipate just as fast.
The fact that most google apps function better on iOS than Pixel phones should be enough to get any sane board to take action. This is without even mentioning the disaster that search has become, or completely dropping the ball on LLMs.
Every large software org eventually becomes Microsoft.
Except when I went to check the Contacts to fix them, they had no birthday at all.
Death by a thousand cuts.
even on apps like google maps. open any place with Chinese chars in the name. toggle the app switcher carousel. try to select the name of the place. only non Chinese text will be copied.
It's crazy to me that it can't just use the accessibility APIs.
It is like that one kid, you know he is trying hard, but he could never make it.
That being said, I no longer have any clue what features exist in Google's ecosystem because their A/B and feature flag rollout system is so fundamentally broken and misused by the product teams that some features literally never arrive for some users until they factory reset their devices, and suddenly they get a fresh set of feature flags. I am still waiting on just about all of the new RCS features Google announced months ago, and just yesterday I spent a bunch of time trying to figure out why my partner's devices never received cross device timers while all of my devices do. That feature was launched in May of last year. My partner does a lot of the cooking, and our kitchen Nest Hub usually handles the timers. But only my phone and tablet receive it's timer status...
I say misused above, because it strongly appears that they announce features before completion and then use an extremely shallow rollout to finish the features with the cover of "finding bugs". Is it actually released if the only Google accounts that receive the flags are the dev and product team members?
Google, you've perverted a fantastic engineering practice into a broken mess that reduces your user's confidence in your product and breeds confusion that hurts your brand.
I personally much prefer Apple's approach where generally features are identified by specific version numbers. If you have the version or newer, you have it. Certainly a bit more difficult to do capacity planning for the backend engineers but absolutely worthwhile for users.
Here is an article with complaints about Google travel. https://skift.com/2020/10/20/u-s-antitrust-lawsuit-faults-go...
I don't know that this is the reason people are noticing less cross over but DMA certainly makes it harder.
I feel like the same can be said for Auto-Correct, also from about the same era. How is it that with all of the edge processing it's still the same old miserable experience? It seemed to me that it actually got worse.
Maybe it was all the kerfuffle about AI ethics that nobbled them for a while.
Both the doomers and e/accs can be wrong. "Today is the worst that AI will be" isn't a foregone conclusion. There aren't many domains of software that haven't degraded in quality, and even functionality, if at least a little bit; and there is no evidence that the same won't happen to AI. Technologists say AGI will happen; but why this happens is really outside the hands of the technologists, somewhat more in the hands of the capitalists, but really outside of anyone's hands and in the capitalist system itself; it can't be stopped.
- Using your voice to send an email, video or audio message. You can still make calls and send text messages - Rescheduling an event in Google Calendar with your voice. You can still schedule a new event. - Asking to take certain actions by voice, such as send a payment, make a reservation, or post to social media. You can still ask Assistant to open your installed apps. - Asking to meditate with Calm. You can still ask for meditation options with media providers such as YouTube.
All of these seem to fall under the umbrella of "features that actually make the assistant an assistant"/connecting the assistant to other apps, which I imagine is exactly the opposite of where the Assistant trend is going, especially with LLMs. Just speaking to a device about which action you want to take and not needing to think which app you need to open and navigate feels like the UX of the future, whatever this is seems like the opposite.
Assistant has trained our household into thinking that its pretty limited in what it can do so everyone in our home only uses it for basic things. Since Google's assistant can't accomplish more than most basic tasks people don't use more than the basic tasks. No one wants to learn the appropriate subset of English to speak the Google assistant dialect. You really need the voice assistants to be reliable and basic or extremely capable. There is no real middle ground here for most users.
Hopefully once LLMs get more integrated with voice assistants we will move more towards the extremely capable side of the spectrum.
The LLM, having been pre trained on what to do, calls the API to turn on the lights.
I'd already noticed my new phone (pixel 8, I fucking hate it) had trouble setting a simple timer with a voice command ( last time I tried it gave me a shitty youtube video result of a 10 minute timer!?!), but this list of removed features goes further, removing other basic functionality from voice assistance.
If all a voice assistant can do is google my "command" and read me the results, that's a terrible experience. The whole selling point (to me) is the ability to do something magic with what I'm asking.
How much does this cost to maintain? A stopwatch? They’re axing a STOPWATCH?!
This is embarrassing.
What do you do if you have statistics saying that, say, a feature is used by a small fraction of users, and 40% of the activations of the feature were in fact not what the user wants, and instead were misunderstandings of requests for something else?
Ambiguity is a massive problem.
But overall I don't like it.
I don't like how it handles notifications, it seems to hide most of them, and the ones it does show it doesn't show in a useful way compared to my old phone. What exactly is different I can't say, but it's weird.
It also handles permissions differently. Some apps when I disable certain permissons get stuck in a permission loop because they can't cope with the idea of only having a particular permission denied. I never had this problem with my old phone despite it also having granular permissions. Seems to be related to a change in how notification permissions are handled exactly.
I don't like how I can't quickly change the ringer volume. It used to be a separate slider when I hit the volume on my old phone. Changing now is a gamble and finding how to get it to vibrate mode seems beyond me.
1Password used to allow biometric unlock on my old phone. I can't get that to work on my new phone so have to type the master password (extremely long password) on a mobile device, which makes it useless to me. I'm not going to type my master password on a phone keypad in public, that's asking for it to leak.
The fingerprint sensor is okay, but I preferred the below-screen sensor on my old phone.
It's fiddly to get back to the home screen. On my old phone if I scanned my thumbprint it would go back to home. On the pixel it's a finicky swipe-up-from-the-bottom thing which sometimes goes home and sometimes goes to an app-switcher.
Setting up the phone required giving google rather more permissions than I was comfortable with, but denying them locked out key features of the phone so I felt forced to accept.
The pixel 8 is also slightly smaller than my old phone, though of course I knew that when I bought it. I didn't realise how much 0.1inches was going to feel like a major loss. I didn't want to go over-sized with the pixel 8 pro though.
Overall a downgrade and really horrible experience compared to my old phone, which I only replaced because it was so old and stopped getting regular updates. It was a Huawei P20 pro and in terms of general use still feels just as fast as it did when it was new.
The only feature I like that the pixel 8 has which my old phone doesn't is the wireless charging. That's a game changer when it comes to preventing wires wearing out.
If an app's notification is set to "silent", it'll push it below a certain divider line, and some notifications may be hidden while in DND mode, and then revealed when out of DND, which is a disjointed experience and not the best, but _is_ configurable.
>I don't like how I can't quickly change the ringer volume. It used to be a separate slider when I hit the volume on my old phone. Changing now is a gamble and
Yeah, this used to be easier. Now you have to click the ellipsis at the bottom of the volume slider to open up the other volumes, but it's clearly labeled.
>finding how to get it to vibrate mode seems beyond me.
This is simple: it's the phone icon at the top of the volume interface when you change volume.
>1Password used to allow biometric unlock on my old phone.
This seems to work for me on my P8Pro
>It's fiddly to get back to the home screen. On my old phone if I scanned my thumbprint it would go back to home. On the pixel it's a finicky swipe-up-from-the-bottom thing which sometimes goes home and sometimes goes to an app-switcher.
I absolutely hate the gesture navigation but I solved this by sticking with the old school 3-button navigation, which you can change in settings.
On the OS side of things, when switching between activities, very rarely my last activity is to the right in the carousel rather than to the left. Because I can't predict when it happens, I can't mindlessly switch to my previous activity, which is surprisingly annoying.
A YouTube video has the chance to earn them advertising revenue. A 10-minute timer does not.
However, they are clearly missing an opportunity: "Your 10-minute timer, sponsored by Burger King, has finished. Would you like to order a Whopper to be delivered by Uber Eats?"
Reading this list of removals, especially the removal of CALLER ID, on video calls no less, an insane choice - how much could that possibily cost to maintain? - I'm honestly wondering if they're actively trying to drive people into Apple's ecosystem.
Nevertheless I'll never buy a Google product as long as I live. How can I trust them not to completely mothball every useful feature I bought it for?
My biggest issue with Siri is intercom: at home we mostly speak other language than english, and when using intercom Siri often tries to parse what's being recorded, instead of just recording audio sample and send it verbatim.
Something like:
Me - Hey Siri, Intercom
Siri - intercom chime (not 'Aha?' of responding to just Hey Siri)
Me - *bjlkabdgkjhqwruo;fghnasd.mkfnlkjashfjkasdngjkasbdg* (doesn't matter, different language)
Siri (thinking she recognized some word) - Here are the results for what you asked..
Siri is sorry she can't help, misunderstands things, or straight up ignores requests way too often.
I will say that the non-display Google devices are often more helpful than the ones with displays. The displays want to show you stuff, but the non-displays will tend to give you an answer.
e.g. "what time does walmart close" will respond with "there are several stores in your area" and then show you a list of stores with their hours, but the non-display devices will just reference the closest store and tell you when it closes.
"Hey Siri, turn off my alarm" (ignores me)
"Hey Siri, turn off my alarm" (ignores me)
"Hey Siri, I'm going to buy an Android" (turns off alarm, refuses to elaborate)
MKBHD Comparison:https://m.youtube.com/watch?v=Q2MGqmuEdtU
Somewhat old but still holds true imho.
Not that it matters to me because having bought an Amazon Echo a few years back I find these things mostly useless. It basically just does the most basic things a phone does (music, weather, news headlines), but with added friction when it misinterprets you. They are a clear example of something we rush to buy because they are "the future" before realising that the future is just the present but more tedious and with more ads.
Its a shame, its been a LONG time since Google surprised and delighted me like they did many times during the 00s.
Hey Google, remind me when I'm home to ...
"I can't remind you at a place" - pretty sure it used to be able to???
Hey Google, remind me tomorrow to ...
"Ok, what time do you want me to remind you tomorrow?"
Noon
"Ok, I'll remind you at noon today"
???????????????????????????
No longer displaying commute time to work, but you can check it on maps? So, they're removing the assistant part into a search box.
Assistant alienated a lot of other departments from 2016-2020 with their rapid rise and rapid growth in headcount (and ability to push integrations through whether they were a good idea or not). Then they failed to make significant money, right into an economic environment where making money was prioritized much more highly than speculative researchey new markets. These changes are them trying to make themselves indispensable to other PA's metrics, so that when budgets are set, they can argue "You can't cut Assistant, look at the drop in searches & YouTube views that would result in." Same reason why they were hit harder in the layoffs than other more profitable products.
At the same time the same executives did not incorporate LLMs into the Assistant when Google invented the tech.
These layoffs should've fired all those dead-weight executives and the CEO for good measure yet we see the rank and file losing their job
Reminds me of legacy mobile phone makers when the iPhone launched.
Instead this seems to be a list of useful features that are being taken away!
I'm disappointed. Google employ some of the best AI/ML engineers in the world and they are making assistant worse rather than making it a showcase for what LLMs can do.
I expect that a new assistant like product (Google Bard Home or something) will be announced and existing Assistant/Home devices will be deprecated.
Was announced in October, but haven't seen anything about it since.
* Playing and controlling audiobooks on Google Play Books with your voice. You can still cast audiobooks from your mobile device.
* Setting or using media alarms, music alarms, or radio alarms on Google Assistant enabled devices. You can create a custom Routine that has similar behavior or use a standard alarm.
* Accessing or managing your cookbook, transfering recipes from device to device, playing an instructional recipe video, or showing step-by-step recipes. You can use Google Assistant to search for recipes across the web and YouTube.
* Managing a stopwatch on Smart Displays and Speakers. You can still set timers and alarms.
* Using your voice to call a device or broadcast a message to your Google Family Group. You can still broadcast to devices in your home.
* Using your voice to send an email, video or audio message. You can still make calls and send text messages.
* Rescheduling an event in Google Calendar with your voice. You can still schedule a new event.
* Using App Launcher in Google Assistant driving mode on Google Maps to read and send messages, make calls, and control media. You can still use voice control on Google Maps the same way.
* Asking to schedule or hear previously scheduled Family Bell announcements. You can create a custom Routine that has similar behavior.
* Asking to meditate with Calm. You can still ask for meditation options with media providers such as YouTube.
* Voice control for activities will no longer be available on Fitbit Sense and Versa 3 devices. You'll need to use the buttons on your device to start, stop, pause, and resume activities. You can still voice control activities on Pixel Watches.
* Viewing your sleep summaries will only be available on Google Smart Displays. You can still ask for sleep details by voice on third-party smart clocks.
* Calls made from speakers and Smart Displays will not show up with a caller ID unless you’re using Duo.
* Viewing the ambient “Commute to Work” time estimates on Smart Displays. You can still ask for commute times and get directions by voice.
* Checking personal travel itineraries by voice. You can still ask for flight status.
* Asking for information about your contacts. You can still make calls to your contacts.
* Asking to take certain actions by voice, such as send a payment, make a reservation, or post to social media. You can still ask Assistant to open your installed apps.
It is one of the few products I use daily that gets noticeably worse over time. Shopping lists, calendars, reminders, timers, etc use to work pretty well, but now they're absolute trash. There's a running joke in our house that one of us will ask the Google Home speaker something, it responds with either "I don't know", "I can't do that", or some bizarre non-sequitur and then we say, "just ask ChatGPT".
It's good enough for cooking, which is the only time I care about not touching my phone.
I know there are some cool DIY hacker projects out there, maybe I should really get my hands dirty tbh? In that case, does anyone know any good links to things they tried recently? Especially with Whisper.cpp and so on it may not be TOO hard on a Raspberry Pi... right?
Instructions from first search result for "disable google assistant" might work? I found the option buried about 5 menus deep[1], but I left it on since it never triggers for me anyways, probably because I didn't set it up.
[1] Settings -> Google -> Google applications -> Search and assistant -> Google Assistant -> General -> Disable
My phone UI language is not English, the list above are my translations, actual text will likely be different.
https://www.howtogeek.com/741457/how-to-disable-the-google-a... worked for me
Note that there has been a significant layoff days ago in the assistant team to "improve Google Assistant as it explores integrating newer artificial intelligence technology into its products". Probably they're trying to replace the backend for the assistant with some LMM-based models and see those functionalities are the blockers? Of course, this could be (and should be IMO) done in more gradual migrations rather than this sudden deprecation, but who knows what those execs have in their mind.
If I understand Google culture correctly, "promo" is the answer.
Tools that people own-- capital, almost--that work for individual own best-interests.
The Personal Computer revolution was made possible by economies of scale in industrial manufacturing of relatively simple components. Companies made money by selling widgets to as many people as possible.
By definition, AI is only made possible by advances in massive computation capacity, which is easier to achieve and improve with centralized models. This is always going to be the realm of Big Money.
In a way, AI epitomizes the shift from XX-century economics, based on manufacturing, to a XXI-century version based on extracting rent from networked digital subjects.
By what? In the words of Inigo Montoya: "The way you use that word... I do not think it means what you think it means."
Sure, we're kinda-brute-forcing things now, but that's no physical law (nor dictionary definition) declaring that digital assistants will be just captive pen-pal lamp-genies of Big Companies inside their Big Computer Rooms forever.
Perhaps someday you'll be able to grab an off-the-shelf AI core, "flash" its mutable-circuits/firmware/weights with a starting state, and pop it into an appropriate chassis for the task.
AI is a buzzword for "tons of clever statistical software running hot to mimic (the output of) our complex brain processes". Big Money will always do that sort of thing better than the Taiwanese, and there is absolutely no incentive to turn a rent-extracting networked service into a widget-selling proposition, even once the field stops progressing at the current pace. Everything is now networked, so everything can access what will always essentially be "Brain as a Service" (BaaS, surely someone coined this already...).
Will we see specialized-hardware AI tuned for this or that task? Probably. Will we have portable configurable AIs? Possibly. Will either of them be as good as Big Money models, which can continuously feed on subscriptions and web-scraping at massive scale? Very, very unlikely imho.
Since I bought them, there have been almost 0 improvements that I've seen. A lot of potential gone to waste. That they're sunsetting features is not a surprise.
I don't know if Alexa is better, but I always tell my friends to buy those instead of Google Homes.
That's not how two humans work on command / response tasks, say, hanging pictures. You say, "a finger to the left and up" and I move the picture. Silent UI feedback is a curiously unplumbed idea.
Or another one: if you sit at a restaurant and ask the menu, would you rather they tediously recite it? Or hand you a paper menu while maybe giving a couple tips like "the fish pasta is awesome today"?
Supposedly, when you are voice commanding, you do so because you are unable to look at the device and visually confirm that the correct action is being undertaken.
i'm sure this is where they're headed. but what announcement reads as to me is some clever PM at google realized they don't want to announce a new LLM update to the assistant and have to include all the functionality it loses in that same announcment. so they're taking the functionality away now, and in six months they can announce a new assistant that has feature parity with the "old one"
who the hell is using Duo
Even Google product teams can't keep track of Google product changes. This one happened a year ago!?
1. I can see how this is absolutely frustrating for someone who uses one of these commands regularly.
2. At the same time, one of the issues with voice commands relative to say keyboard input is the lack of delineation. If I hit Ctrl + Enter on my keyboard, there's no way for the computer to interpret that any differently. However, that's not the case with voice commands. It needs to keep determining (a) what I said, (b) whether it's a command or just voice, (c) whether it's a command it's supposed to execute and that's before it even figures out waht to execute. But the problem is that as the number of voice commands increases, the number of false "this is a command and not just voice" triggers will increase.
There's probably a genuine benefit for Google's users in the universe of commands being very small as long as the command that's removed is not one they use.
I've noticed several of the options are "you can use a custom routine to do this instead". Maybe Google should have included the custom routine and made it togglable.
Every major company has been working on an AI assistant for the past year. Have any of them launched something useful?
All this is to say I'm fairly late to the voice assistant party. At some point my wife and I decided to give it a good try. Read up on best practices. Figured it'd been something that had been in market a good long time, maybe by now the kinks were worked out.
The search results were hit or miss. It was not clear where tasks were being saved to.
So for the last few years, here are the voice commands we've used:
ok google, what's the weather today
ok google, play white noise
ok google, stop
...and every now and then, we ask it to search for info on a topic and it fails comically. I'm not surprised to see a big shift here.
Oooh. On my Pixel, I do use "OK Google, Turn off the flashlight" - that IS helpful.
Then again, 99% of my Assistant use is kitchen timers and Spotify, and I systematically disable it on my phones. My kids are still salty about Song Quiz going away last year though.
> Managing a stopwatch on Smart Displays and Speakers. You can still set timers and alarms.
You already have a clock app, probably based on the standard Android one, and how many commands are for stopwatches?
> Using your voice to send an email, video or audio message. You can still make calls and send text messages.
What’s the big difference between sending a text message to a phone number via the messaging app, and sending an email via the Gmail app? It’s probably a single API call with a few parameters.
> Rescheduling an event in Google Calendar with your voice. You can still schedule a new event.
> Calls made from speakers and Smart Displays will not show up with a caller ID unless you’re using Duo.
They're going to take this little piece of convenience for me that I always use. Sad :-(
expecting a resurgence of all these natural language assistants.
I was expecting to see the effectiveness of these assistants go through a step change for the better, and use cases exploding.
Expected the established players like Google and Amazon being able to finally justify building so many of them and selling them cheap with and uncertain pay-off all these years.
Alas!
This seems to imply that apps which feature 'app actions' would no longer work properly and will need to be opened via assistant separately...
Sounds like a major regression for those scenarios with needing to issue a separate command to open an app.
I went home from that interview and threw my Google Home device in the trash.
dog barking, baby crying, snoring, glass shattering, smoke alarm beeping, co2 alarm beeping, appliance beeping, running water
My primary use-case is creating reminders. For this I use (of course), the microphone in the search bar. I guess I'll have to find the next best way. Maybe it will end up being better.
I'm down to three gestures with assistant:
- play white noise. 50% success rate. Half the time it plays white stripes or some random death metal, which is great when my kids are trying to sleep.
- "ok google, stop". This works reasonably reliably.
- "I wasn't talking to you". For the 3-6 times per day when it activates for seemingly no reason and where no activation logs are ever created.
If I can find a decent internet-attached photo frame to replace it I'm out.
It feels like 2010's tech that couldn't keep up.
It's like when you get the "we're updating our privacy policy" email
Makes you think.
Presumably Google Assistant is on it's way to being replaced by ChatGPT with plugins, although these would need access to phone APIs to do local stuff.
anyway, we came back home and still don't use siri
We use it quite a lot.
Simple things through voice are good, and a lot better than fiddling with your phone while you're trying to drive.
"Text $WIFE_NAME I'm on my way home"
"Directions to $DESTINATION"
"[Set a] Timer [for] 15 minutes"
"Call $CONTACT" or "Call $PLACE"
etc.
These are usually done while I'm driving or on the move, or otherwise don't feel like scrunching up my thumbs to type or to look at a screen
As always: "Bug fixes and performance improvements". /s
Make it truly useful or get out of my way.