Slashdot Mirror
Ask Slashdot: Is There a Useful Voice-Activated PC? (dailycaring.com)
An anonymous reader writes:
My elderly monther-in-law misses her computer. Her mind is okay, but she cannot use a computer because of her Parkinson's disease.
I am not all that impressed with Amazon Echo. Seems you can ask the Echo for the time of day, or the weather outside, but it will not do anything useful -- like send an email. A voice controlled PC would be great, even if it only did a few simple tasks.
The original submission ends with a question: "Is there such a thing?" So leave your best thoughts and suggestions in the comments. Is there a useful voice-activated PC?
I am not all that impressed with Amazon Echo. Seems you can ask the Echo for the time of day, or the weather outside, but it will not do anything useful -- like send an email. A voice controlled PC would be great, even if it only did a few simple tasks.
The original submission ends with a question: "Is there such a thing?" So leave your best thoughts and suggestions in the comments. Is there a useful voice-activated PC?
Ok Goggle, nevermind
Windows since XP and MacOS since like... 9.0 have included voice assistive technologies.
You just have to turn them on.
If the OS's built in speech recognition and control don't do what you want, buy a copy of Dragon NaturallySpeaking.
An iPad with proper accessibility settings and Siri voice recognition.
Iâ(TM)m not a native speaker and Siri does understand me well enough to write proper emails and messages.
https://www.abilitynet.org.uk/...
Of course not. A normal PC is useful. But if you simply took a normal PC, and added a microphone and some voice activation software, it would immediately cease to be useful. Hence there is no useful voice-activated PC.
I know you need to give access to millions of sensitive data in order to use Cortana (Windows 10). I have not had much success on either Cortana ir OK Google as I am not an English speaker. I know other people had success with Siri.
Because no voice activated PC ever didn't make the user angry by doing the wrong f--king thing at the wrong time. On top of this you're probably going to force windows on her as well.
At least she'll get a good workout throwing the f.?king thing out of the window when it really gets on her nerves.
And I've used them successfully, too. To control the PC and to write documents.
They are called iPads.
Not a direct answer, but perhaps helpful -- there is some promising work being done with tremor compensation/cancellation technology. Strap on a bracelet with a type of vibrator attached and it can stabilize your hand movements, kind of like camera stabilization does for taking pictures.
https://www.digitaltrends.com/cool-tech/projects-backed-by-google-and-microsoft-are-tackling-parkinsons-disease/
I think that does what you are looking for.
Having spent a certain amount of time with disability offices I suggest that you go and find a local Blind Association and see if they can allow her to test out some of the software written for blind people. Jaws is a software that allows you too hear in one ear what's on the screen and what you're typing in the other ear might be more difficult than your mother-in-law can manage. The other thing is that this will probably help her get access two other entertainments such as audio books and people in a similar condition to commiserate. What you're looking for is probably not an out-of-the-box solution sold at the General Market but something you're going to find it's a little more expensive but there's probably a grant for it.
Life is like untied shoe laces; it always tripping you up and getting in your way.
the voice that your device will pick up, and the video that the camera on your device will pick up, is not there to make your like easier, there will never EVER be a kinnect that you stand in front off to interact in a meaningful way with your your computer, all that tecnology its there to just to spy on you, it has no other purpose and will never ever be more than a gimmick at most
Obviously setting up these devices can take some customization, and bringing up the Amazon Echo, a device not meant to be a computer, while appearing ignorant of the existing voice-assistance technologies in actual computer OSes is a warning sign that you may need real hands-on help.
Not that we won't try here, but I've got to get that advice out of the way.
The answer is simple: The technology has to BE ON YOUR COMPUTER. It has to work offline as well. Otherwise this is just the usual "giving away your most private information for free".
Know a lawyer with a brother with disability. They added https://www.nuance.com/dragon.html
To his PC.
It comes from the same people who make automated lawn mowers.
They're called children.
The overwhelming problem with all of these speech recognition interfaces is that web sites are not coded for key word searches and every website on the planet would most likely screw up the idea of using a keyword search structure. Again it all comes down to language and the fact that the complexity involved in obtaining fine grained results from key words in combinations interpolated by a computer is an enormous task that is fraught with the possibility of error.
This message was not sent from an iPhone because Peter Sellers really was a deviated prevert without a dime for the call
Nuance's Dragon NaturallySpeaking is advertised as such an accessibility solution. IMO, a programmer might be helpful in case a little coding and registry manipulation is needed.
I had a similar problem with an aging parent with RA that was so bad she could not type. I asked her what she wanted the computer to do for her. She gave me a list of 7-8 things (search the web, write an email, open her facebook page and weather were the top ones) so I downloaded https://mega-voice-command.com/ and went to work. It took about an hour to set it up and get it to respond to her the way she wanted. That was 8 months ago and now we have about 80 custom commands in it and I just put another mic in her living area so she doesn't have to get up to use it. It even works if the 'net is down because it is a local DB and not an online one like Alexa and it takes up little room, unlike Dragon.
I get a LOT of emails and facebook posts from gams now!
And remains free for any person with accessibility needs. I'm not even talking about Cortana though you should consider it. Just because /. hate it for irrational reasons doesn't mean that she should have to struggle with an ancient and sub par experience
On the MacOS, iOS and tvOS, all from Apple, there is an accessibility service called Switch Control. Similar software exists for other operating systems. Switch Control along with the built in dictation service provides a means for those with very limited mobility to operate a computer or similar device.
When connected to a bluetooth switch such as those made by Ablenet (www.ablenetinc.com) the whole interface can be control from a single button which can be bushed by the disabled person. Other switches work by means of puffing into a tube or tracking head movements. People with very limited movement are able to employ this to use computers and there are a number of videos of such use on YouTube. Dictation, now built into Windows and MacOS, iOS, tvOS and Android could be used to provide easy text entry. Combining these and the voice control functions of modern operating system should provide her with the tools needed to again access her computer.
X10 (industry standard) https://en.wikipedia.org/wiki/...
With a home computer, the home automation kit and no internet many tasks can be automated using voice.
Generations of people in the USA used their DOS and Windows voice control to automate their lights, thermostats, alarms, some appliances.
Such systems did the task asked for when told. No need for a robot buddy or todays ad and internet connected systems.
Domestic spying is now "Benign Information Gathering"
The big tech companies are pushing the digital assistant technology hard, but the truth is that, as things currently stand, the usefulness of such assistants is extremely limited. They are able to answer very specific, domain-limited questions which, for the most part, you could answer yourself probably just as quickly, if not more so, from your keyboard. Other than that, they do very little that is really useful. My Alexa device can turn lights on and off - which I can do faster by throwing a switch. I guess it is useful when you want to do it remotely. OK. As for the weather, both Alexa and Google (whatever it is called today) tell you the weather forecast all right - but their understanding is nonexistent. Tell them explicitly NOT to give you the weather forecast, and they will give you the weather forecast. Or say some nonsense that includes the word "weather", and they will give you the weather forecast. Try to get them to do something slightly complicated or ambiguous, the kind of thing that a child would accomplish without any problems, and they will consistently fail. Are they useful? In very, very limited ways. Are they worth the while? For grins and giggles, yes. For just about anything serious, what you would expect from a human assistant, no. Maybe in 10 years time. However, bearing in mind the track record of the AI community, whereby they have consistently gotten carried away in their projections, I remain skeptical.
Mac OS X has had this feature for nearly 20 years.
Step 1: "Computer, open facebook.com"
Step 2: ???
Step 3: Profit!
https://www.youtube.com/watch?v=lCDU8r7Tcfk
Had me in stitches XD
My mother also has Parkinson's - and a Chromebook.
She can still use the touchpad, but typing is no longer possible. So she dictates her search terms with "OK Google".
The voice recognition expands to general text input, but dictating e-mails takes a bit of goodwill and some time to get acquainted with.
OK Google and Chrome Browser have been mentioned, but the question asks for a computer.
No
It little behooves the best of us to comment on the rest of us.
Several are useful (saving time or complexity), but mostly in very limited spaces. Google's is useful when you don't want or have the free hands for typing a query into your phone. Echo is useful for playing background music. Even Kinect's can be useful for commands on a TV. If you want a fully functioning assistant, hire one; AI isn't there yet.
Real lawyers write in C++
You can train your PC to do that. I use dragon naturally speaking with dragonfly python scripting. There are many python libraries which are built on dragonfly, which do the job.
Caster is a good one. Check these tutorials.
You will need to spend some time to understand how things work and train your mother-in-law to use the customized voice commands.
Search for "dragon naturally speaking demo" on YouTube to see what others do.
Basically it would have to be near human voice recognition if it were implimented with humans.
Frankly, I've never liked voice interfaces. They don't stop and start recognizing when they should execute properly. And the accuracy of command execution is poor. And the speed of data entry or command entry is slow.
I can see some uses as a back up interface that would ignore commands unless certain conditions were met... and was only used for a few things. I don't see the use for it outside of that until the voice recognition AI gets pretty near Rosie the Robot.
I've decided to stop wasting my time responding to AC trolls/sockpuppets... so if you want a response from me... login.
I have a friend who is a quadriplegic and lives in an electric chair. He is also a software engineer and very active on Facebook. Last time I knew the details of his setup he was using Dragon, I believe. As I understand it, it's fully customizable, i.e. you get to tie particular voice commands that you choose to particular actions, widgets, keystrokes, etc.
It took him a year or three to get it all customized to his liking for everything, but at this point he basically rolls around and uses the laptop attached to the deck on his chair in front of him nonstop. He's got a bunch of IoT/smart home stuff set up at home and in his office as well, he provided directions and his wife set it all up under his supervision.
The result is that he basically has a workable voice interface to the Internet, his IDE, Windows, and also most of his immediate physical surroundings, so that he lives a fairly normal life, apart from bodily functions and eating, which he obviously needs help with. But most everything else, from rolling around/chair control to lights and blinds and doors and windows and locks to television and computer and work he does by himself, without any movement in his limbs, using voice.
All off-the-shelf stuff as far as I know, they're very middle class and bog standard insurance, no huge budget, just a lot of his expertise and his wife's hands to set it all up over the years.
STOP . AMERICA . NOW
I sympathize. There aren't a lot of options for people with disability or disease conditions that impair their mobility who still want to interact with their PC. Most speech products are transcription products, and then there's Cortana which I've never really used. But there is one product I can recommend, called "tazti" speech recognition. I know a number of disabled people who use tazti to control their PC and play video games. It's *not* dictation software like Dragon, only command and control for the PC. You create custom speech commands and then specify what they do - Basic PC functions, mouse and keyboard, sending keystrokes, running programs or opening files or bat files etc. It doesn't control mouse movement but it lets you click and double click the mouse. And you can group commands together for profiles - for example you can have a different custom profile for each program or game, and you turn on and off the profiles using your voice. And you can download some pre-build profiles. The other big thing about tazti is that's it highly customizable. You an create Macro commands activated by voice for games or programs as well. And they have good customer support as well. It's really neat, and while it may not provide 100% coverage for your mom, it covers a whole lot of ground and may be really helpful to her, a big improvement over what what she currently has. Anyway, like I said, I know disabled gamers who use it to play games, and people with disabilities who use it to interact with their PC. It's worth checking out. The URL is https://tazti.com/.
Contact your local association for the blind and visually impaired. They can probably suggest voice controlled software you can use. Even though your mother in law is not blind, I think many features of PC controlling software made for the blind would work for her.
not /. material. post this on yahoo
stoooopid
I was shocked that neither Amazon Echo nor Google Home can take voice dictation! They've got speech recognition, so surely they should be able to at least convert that to text and have it read back in Alexa/Google's voice later on at a bare minimum (this assumes it can't record audio and play it back later...which would be nice too).
Before you mention EchoScribe, that was a total hack involving an external microphone amongst other kludges (and the website is sneaky not mentioning the kludges up front)!
...and a lot of laptops (and maybe Chromebooks or boxes) have mics on them. It should be possible to use it to compose emails and such. I know Android has a lot of voice capabilities with "OK, Google" and I assume it extends to ChromeOS at the very least (and might include the Chrome browser).
I have to say if google mini is anything to go by, we are at the stage where if you accept account linkage and storing your requests by an ad company, we can reach parity with a bad 80s style text adventure, think "use lock" "turn key" "open the damn door you useless pos"! I use plex with phlex tv written by a single dev afaik which does a better job of finding correct songs from my library than google in google music with the same songs, also google hobble using assistant on tablets and have no viable pc version, also cortana looks as though it wants to key log before operating! I seriously hope mozilla and baidus open source speech recognitions pulls us away from the crap Ive seen so far!
Android (with Google) has excellent voice control. At least for what I have tried.
"Email to {name of my contact}" opens a blank email (in Gmail) and then asks me "what's the message"? Say the message and that becomes the body.
Then it asks "do you want to send this?" and you answer yes or no. It all works fine.
Haven't tried too much else, but "play {artist} on Spotify" starts Spotify and plays artist radio for that artist. I would imagine many other apps work similarly.
There are other assistive technologies other than speech control -- e.g. keyboard with keyguards. Essentially a plastic overlay for the keyboard where you can rest your hands directly on the keyboard itself without accidentally pressing any, but it has holes through which to actually reach individual keys.
Windows itself has accessbility features as well, such as StickyKeys (press modifier keys separately instead of simultaneously) and FilterKeys (ignore rapid succession keypresses)
There's other assistive technologies for people with Parkinson's and/or tremors as well, quick google search shows dozens of different approaches. Don't ignore voice control of course, but don't think it's the only option.
Hey Siri, pour hot grits down my pants
It works even when the PC is turned off https://www.blackhat.com/docs/eu-17/materials/eu-17-Goryachy-How-To-Hack-A-Turned-Off-Computer-Or-Running-Unsigned-Code-In-Intel-Management-Engine-wp.pdf
Voice as an option on Macs must have been around longer than we think. Back in '86 when I was working at Plexicorp, a visiting professor from Edinburgh University tried to use my boss's Mac as if it had a voice interface. He ended up falling back to the keyboard. He seem very bemused to have to use such a primitive interface. The prof seemed to really know about engineering.
I have Parkinson's disease. It affects people differently. I use speech recognition productively, but in advanced cases, the speech of the person with PD can be adversely affected. If this is the case, computer speech recognition is not going to help.
As others have mentioned, look into Dragon. I spent some time with an accessibility researcher discussing this; he himself is confined to a wheelchair and can only control the chair's small joystick, which doubles as a bluetooth mouse. For people with motor issues of all sorts, it's common to either use some kind of control stick or voice software, and Dragon is apparently one of the easiest, especially once it's personalized with various control macros to trigger your own most common computing tasks.
JAWS is another common accessibility control software, and there are a few smaller open source options as well.
My father â" a true rocket scientist, programmer, and all-around nerd â" lived with Parkinsonâ(TM)s for 16 years. When typing started to become difficult, he used Dragon, as others here have mentioned. After getting his first Deep Brain Stimulator, his tremors became controlled enough that he could resume almost full normal use of the computer, falling back to voice recognition only when needing to input large quantities of text. Of course, the disease is progressive, and eventually using a computer became too difficult either way, mainly due to difficulty maintaining concentration. A voice interface may make some things easier, but it also makes many things more difficult. As far as I know, there is no comprehensive voice interface system that addresses all the needs of a typical computer user, let alone one suffering from a neuro-degenerative disease.
I asked my computer. It said "No".
(by advance, please excuse my language as I'm not a native english speaker)
They (Tobii) have a lot of products designed for disabled people and they seem to have a great product line, not just for FPS gamers ;)
For your mother-in-law a combination of the TobiiEye hardware with their Communicator software seems to be the real deal.
If she already has a PC, then maybe adding this hardware could fill her computer needs :
https://www.tobiidynavox.com/fr/devices/commandes-oculaires/pceye-plus/
But the Communicator software seems really cool :
https://www.tobiidynavox.com/fr/software-apps/logiciel-windows/communicator-5/?MarketPopupClicked=true
They may be only Windows 10 compatible, no linux or mac ...
As a side note, they also have tablets : ...
https://www.tobiidynavox.com/fr/devices/commandes-oculaires/i-15/
But I don't know if the software is actively maintained
Hope that helps ...
As I said I'm not a native english speaker, so I linked to the french version of the above pages.
Here are the english ones :
Communicator :
https://www.tobiidynavox.com/software/windows-software/communicator-5/?MarketPopupClicked=true
PCEye :
https://www.tobiidynavox.com/devices/Eye-Gaze-Devices/pceye-plus/