My father has hearing aids and in some environments, it doesn't work well (directional vs omnidirectional ... but directional requires the person to directly address him while omnidirectional tends to pick up too much background clutter). In doctors offices (for example) he has handed his phone in dictation mode to the receptionist and asked them to speak to it and then he reads the transcription and responds... and this repeats. In a restaurant environment, sometimes another person orders for him.
This sort of thing could be useful in such cases.
Or are they the directional ones you were referring to?
One of the features that the app can control is the "environment". When out and about it's set to omnidirectional so that someone next to him doesn't need to be talking directly to him, however that picks up a lot of background and when in a building that causes.
He's also wired up their house with an audio loop. There's a microphone at my mother's craft desk that goes directly to the hearing loop and he can hear it anywhere in the main rooms of the house.
However, going out to places where he needs to be able to talk to people with less assistive technology available, things like the dictation mode on the phone area used. Also, sometimes he'll miss words (the start of a sentence until he realizes someone is talking to him, turns his head and the directional hearing aid mode gets them in focus). Being able to review the last 15 seconds of what was said on his iWatch would a technology that he would use.
It's built into the meeting software I use and probably in most mainstream ones.
I tried OS transcription but those don't split out who's talking (since it's a single audio stream) so much less useful.
Anyways all that to say that at least in online meetings this problem is solved and addressing this problem in a stack that doesn't have meeting context is going to result in a worse experience.
That’s what the wrist computer is doing.
Same with the "airpod live translation" thing, they probably spent more time implementing/testing it than the feature has been used in real life
It’s hard for me to imagine not wanting language barriers to disappear.
Did you mean to use another word here? I mean, the vast majority of people are still on Earth so many things are geocentric and that's rarely considered a problem.
Hazarding a guess at egocentric considering it's just two characters next to each other swapped, and happens to fit.
So either way, wrong word.
Interactiving with people is so pre-covid!
You'd be shocked what the suite of sensors in the smartphone 99% of the population has in their pocket can do.
This is basically just "always on" recording/transcribing that people have to intentionally remember to turn off, that's a pretty big difference.
This featuring is undoubtedly to compete with, and crush, all of those stand-alone AI wearables that do the same thing. All those products failed, because people either didn’t find it that useful, or the creep factor was too high.
I could see some scenarios where it might be useful. Being at a conference, for example, and having a summary of everything I heard in the various presentations. But at the same time, it feels like the people being recorded should have knowledge of, and give consent to, the recording.
At least with the Apple one they talked about how it works and it’s not some random product off Amazon doing god knows what.
My dad is probably the most likely person in my life to use something like this. I could see him sending me the summary of our dinner conversation or something, which would be very weird and off-putting. He has been very concerned about his memory as he ages, so I could see him using something like this to help with age related memory issues.
It likely stretches the definitions compared to what most people who do not want to be recorded think the laws should prevent, but on the other hand at least it couldn’t be used as legal evidence. It sounds intended as a memory aid, where passing a transcript for gossip purposes will hopefully be met with automatic skepticism as something that isn’t authenticated in any way and could be made up entirely.
They probably spent enormous amounts of time weighing how much legal jeopardy it would create for their customers.
The world’s fastest smartphone typist can do what the mic icon does. In, say, court, “look I typed this” & “look this was dictated” would seem to be just as trustworthy since either can be edited after the fact. And someone could type and say it was transcribed, or transcribe and say it was typed. Whereas the law doesn’t want us voice recording and having the actual (forensically provable? w/the right budget?) record, unless everyone agreed.
A short buffer is common with wake word audio processing.
Regardless, you'll be mortified if you search the app store for software or Amazon for hardware devices that do this! You can still pick up voice recorders at Staples, actually (I was surprised to see they still sell them, and they have weeks of recording time now, along with transcription software).
I’m curious if the lack of speaker identification is a privacy thing, or a technical limitation.
It looks like it uses PCC, so it may run into usage limits
“Siri Recap”
> Siri Recap and Live Rewind help jog your memory, not provide a detailed record. Siri Recap doesn't produce a transcript. It generates a brief, high-level summary — comparable to notes you might jot down after a conversation. Live Rewind shows a text snippet of what was just said in the last 15 seconds, so you can catch something you missed.
I have real respect for Apple's "Private Cloud Compute" architecture [0], I am not a hater, just trying to understand.
Are these the dividends Apple is getting from doing this truly impressive work?
[0] https://news.ycombinator.com/item?id=49427928 - https://security.apple.com/blog/private-cloud-compute/
The regulatory workaround is an "un-mutable" audio notification of some sort. This might be a very notable sound in the near future. I am curious what the chosen waveform might be:
https://support.apple.com/en-us/148354#:~:text=Audible,activ...
I am not a lawyer but I do not believe that is true in any state except Nevada. Two-party consent almost always includes in person conversations. California Code, Penal Code 632:
>A person who, intentionally and without the consent of all parties to a confidential communication, uses an electronic amplifying or recording device to eavesdrop upon or record the confidential communication, whether the communication is carried on among the parties in the presence of one another or by means of a telegraph, telephone, or other device, except a radio, shall be punished by a fine not exceeding two thousand five hundred dollars ($2,500) per violation, or imprisonment in a county jail not exceeding one year
https://support.apple.com/en-us/148354#:~:text=app%2E-,How,s...
> Audio Intelligence features are private by design
> a dedicated hardware-isolated compartment that processes audio in complete isolation from the rest of the system, then immediately deletes it
> Users are in control and [...] can decide exactly where and when Siri Recap takes notes, and can turn Siri Recap on or off at any time
> Siri Recap produces a brief, high-level summary of conversations — not a transcript — and is designed to exclude sensitive information like financial information or government-assigned identifiers
It’s also much less effective at recording people speaking for long durations than the audio recording feature that every smartphone for the last 20 years has supported.
There is no privacy left. And to be honest, people worshipping their phones all day but complaining about privacy is the biggest WTF I have. This doesn't match.
Also I am a Mathematician and know statistics very well. The person, who is not on Facebook, but on WhatsApp, who works at X and lives in Y - we got you my friend.
You moved away? No problem! Someone photographed you casually, you weren't even the main character, we ignored you, and send it in a group chat you know nothing about. Well, we know all the others and can relate to anyway. And then? Well AI can very well predict you aging - easy. We have you already. Sleep well and dream of being unknown. Because... How do you know what you don't know?
I gave up on privacy because there is none anymore. Selling and publication of your data is a whole different story. And that's what Apple means by circumventing the rest.
What people around me? I exist in my own bubble. It's hip and rebranded selfmaxxing or something these days
Get off my lawn !1
Just because your voice was recorded doesn’t necessarily mean it can be played back and/or “stolen”.
Your soul won’t be sucked out through Apple Watches.
If someone is following you around within earshot and fiddling with... a notepad and pencil or a watch or a phone or a tape recorder, then yes, you have reason to be upset. I agree.
The ability of the Apple Watch to do what has always been possible doesn't materially change your privacy. [1]
Fundamentally, this is an accessibility feature, and Apple -- and other vendors -- have already shipped several similar ones. Were you equally upset by those too? AirProds already 'record' all audio around them! They can even be used as listening devices! https://support.apple.com/en-au/102479
This reminds me of parents getting hysterical when someone pulls out a DSLR at a kid's soccer game, while standing next to half a dozen parents who are all taking pictures with the phones and immediately uploading them to Meta.
[1] There are valid arguments to be made that this will be normalised and cloned... badly by Chinese and South Korean vendors without the privacy safeguards, at which point it is "too late" to put the genie back in the bottle.
And I don't think we can compare this to weirdos being able to record other people, this is about when my friends come over and I have to wonder if they think it's normal to have running since "it's technically not recording" or "the summary is anonymized. It's normalizing behavior that's just a little weird, like people who use those ai recorders. I fail to see how that's not a violation of my privacy in some way.
Also you don't need to point out stupid gatchas about airpods, don't be an ass.
I respect that you believe that. I do not. It touches up against my basic relationship to reality and existence. To me, there’s a difference between a person and a machine. And between a meeting-of-the-minds-in-the-moment and an archival record.
To me, even the act of processing me through something-other-than-your-human-mind feels violative—essentializing and reductive. For me it’s wrapped up in ideas about attachment, and illusions of permanence, and attention and presence and the extra ineffable elements of human contact that unfold at levels other than the words and an exchange being about much more than the “information content.”
You don’t have to agree with me, but it’s frustrating for my conversation partner to take what I bring to a conversation while demanding to also take more than I’m willing to offer.
You’re free to allow people to record you, or analyze you, or machine-summarize you, or livestream you for the judgment and entertainment of the internet-dwelling masses, or whatever you’d like.
And I’m free to say: do not record me. Ever. Even on a rolling buffer, even “just for note-taking.” Even if it “only lives” some specific digital place. Unless I specifically say it’s OK, ahead of time; and we agree on how specifically the processed-me-byproduct will be used.
I’m not unreasonable! Sometimes the answer is yes! Is it for accessibility? Of course, go for it! Thank you for asking. Do we want to make a statement for the record? Great! But ask.
Funnily enough the “soul sucked out” trope occurred to me too, but it seems like we take different lessons from it. For me, it cues stories about a bunch of economically powerful people inserting themselves into places and situations where other people lived. Using their new camera technology to exploit those people’s images and likenesses, and justifying it by rolling their eyes together at how superstitious and backward and dumb those people were for not liking it. As if it were impossible for somebody to understand the thing and still not want it used on them.
And thus figuring their superiority gave them permission to “educate” the people whose likenesses they sought to exploit. And if the subjects still didn’t like it, it must be their shortcoming (since all right-thinking people obviously agree about matters of the soul and of cameras).
So it was OK to disregard their subjects’ beliefs, and their subjects’ dignity. The camera-pushers were entitled, by dint of their superior certainty, to treat people like specimens.
I agree there’s a parallel, but I think the people who were in the wrong were the people forcing the cameras on people who didn’t want to be involved.
[ETA, for @cindyllm: Your provocations are sublime, and I'm grateful that you continue to confound those of us who walk with the dead. You keep me off balance like nobody else. Is that your LinkedIn, or would I be spamming a random Canadian person?]
Running a voice memory all day long is one thing. Pressing a button that automatically summarizes every conversation you'll ever have is a step change.
I don't appreciate Apple normalizing the already creepy ai note takers that exist and increasing the odds that someone will use them in casual or intimate conversations.
Yeah, I just decided the entire Apple Watch brand is suddenly a creeper device, so I'm not going to wear mine out, nor otherwise around other people.
I don't want to endorse creeper devices by appearing to wear one, and I don't want to be insensitive to people who might wonder whether I'm "note taking" everything we say.
Also, I got an iPhone specifically to use the Apple Watch awhile, so I'm probably switching back to GrapheneOS sooner than expected.
Remember, this is all running via local models on-device.
This is how Siri/openclaw can hear you make plans for lunch, then add it to the calendar before you open your phone.
Miniature cameras on your body, and summaries of every conversation you have which you can store permanently, are concepts from either spy movies or weird sci-fi. I think it's a big mistake to see this as a continuation and not as a very different trend.
We may have to accept that Big Tech people are weird and cannot be trusted with any of this. See the Valley people now feeding all their personal conversations to agents. They are creepy and they are creating products in their own image.
Cameras were invented before big tech. Big tech took cameras and made them better and more ubiquitous. Microphones were invented before big tech. Big tech is taking them and making them better and more ubiquitous.
To be clear, I agree that the leaders of these companies are weird and untrustworthy. I'm arguing that the utility they will offer the world is worth far more to most people than the privacy concerns. Because of this, most people will happily adopt this technology.
Apple details how Apple Watch’s new Audio Intelligence features work in privacy paper https://news.ycombinator.com/item?id=49635185
Apple Doesn’t Want You to Worry About the New Apple Watch’s Listening Features https://news.ycombinator.com/item?id=49634947
I always knew this was coming eventually, with storage and compute getting so cheap, eventually every record can and then will be saved for security / convenience reasons. Every word uttered, every emotion shown on your face, every URL visited, every view you laid your eyes on.
I will miss the days that things could happen and be forgotten.
No doubt this stuff is all very useful, but I'd rather we skip this line of technology.
The degree to which people are going apeshit over this is absolutely absurd. People need to get a grip.
Always listening in that the owner could run an application that could be recording them and you on the other hand has been a possibility since pretty much day one.
Apple Watch Spy
Apple Watch Tattletale
Apple Informer
Watchholes
iSnitches
Ted Chiang (author of Arrival) wrote about this
I don't want an always-listening watch. I don't want earbuds with cameras. I don't want people around me to have those things either. I hate it.
People love to get worked up over technology that is actually private and secure.
Meanwhile they’ll use Google and Facebook without a second thought.
You’re sounding… hysterical.
Even if you regularly say things that must never be tied back to you, like.. you are in the mafia or you’re an actual spy or whatever, and every single person around you has this on their wrist listening to every word you say all of the time… Nothing changes for you.
The users of the watches still have to take action to use the audio, which they too would have heard anyway, and could have transcribed and texted to someone or written down.
Oh… wait… are you MIB? Are you concerned that your memory eraser wand won’t wipe Apple Watch recordings of the last 15 seconds, allowing the public to finally know about aliens amongst us?
I know that sounds silly, but unless you can erase the memories stored in human brains, then you must realise that everything you say in the presence of meat computers with audio listening devices attached (ears) can be recorded (memorised).
I would hope to not be friends with such a fucking weird person tbh. But I guess if it's automated into a sexy device it's fine.
This isn’t on all the time. It’s on in 15 second increments and clearly chimes when enabled. This is literally just the equivalent of saying “one sec while I get a pen to write this down” and then taking a note.
The difference is extremely obvious, to the point that you even typed out one of the differences!
If you read that quote carefully you'll also notice it's only the audio that's ever mentioned as inaccessible to apple etc but never mentions anything about the actual transcript, wanna bet that goes into their little advertisement venture?
“Audio from the microphone enters the Secure Exclave of Apple Watch, where it is initially processed for speech, sounds, or music, without transcribing or storing the raw audio. This buffer is a continuously overwritten stream that exists only within the protected hardware and never creates an audio recording”
https://www.apple.com/privacy/docs/Audio_Intelligence_Privac...
IDK about you but most of the shit I say, I'd be happy for the hearer's assistant to be listening and potentially adding the things we agreed on to their to-do lists or calendars.