Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

25 Years In Speech Technology and I still don’t talk to my computer

matthewkaras.medium.com

Post reply on HN
Print view View original

111–120 of 296 posts

Re: 25 Years In Speech Technology and I still don’t talk to my computer

#111

Post by crazygringo »

For most short interactions, the mouse/trackpad/finger is simply faster.

Now for long-form typing, I'd love to use dictation and sometimes do for taking down short thoughts I e-mail myself from my iPhone.

But the problem is not just that it still makes tons of mistakes. (Probably a quarter of my notes-to-self involve errors so big it's even impossible for me to later figure out what I even meant by trying to sound it out phonetically.)

The problem is that I can't correct those mistakes using voice. There's no way to say "pause, correct affect to effect" or anything like that.

Even more maddeningly, the words keep changing in real time. Sometimes I'll utter a sentence it gets right, then it "re-analyzes" it and completely messes half of it up.

I just wish there were a kind of dictation where I could say a phrase, pause, see if it's right (and it wouldn't change after), say the wrong part with a kind of emphasis that lets the system know I'm issuing a correction, the system would look for the next most probable alternative, repeat as desired. Then I could actually dictate successfully.

This UX where the words are always changing back and forth according to updated statistical probabilities, even as long as 15 seconds after I said them, and where there's no ability to go back and correct them with voice... it's just so so dumb.

The problem isn't voice recognition anymore. It's voice correction.

Re: 25 Years In Speech Technology and I still don’t talk to my computer

#112

Post by 542354234235 »

memling wrote: post #36
wolco2 wrote: post #21

The problem is we (as in hn) can type faster than we can think of the words to describe. My wife always talks to google. It seems backwards to me. Next generation should skip speaking and wireless read thoughts. Language/speaking is a bottleneck.

> The problem is we (as in hn) can type faster than we can think of the words to describe. Do you suppose it's possible that speed isn't really what counts? In the era of typewriters, authors like C. S. Lewis opined that typing obscured their thinking because it was too fast . It didn't let them savor the words effectively. Maybe what we really need is to slow down?

> Maybe what we really need is to slow down?

In some cases, we should slow down. But in general, I disagree. Especially because of the way in which we use technology now. Our smart phones are becoming a bit of an external brain to us. It holds contacts, conversations, searches, notes, musings, etc. We use it to recall a fact or answer a question without really even thinking about it. Maybe what we really need is to slow don’t think it is only a problem now because smart phones and the like aren’t really designed to improve our lives, but as ad delivery platforms to manipulate us into buying products and services we probably don’t need. I would love better tech that helped offload mental tasks my human brain isn’t great at, but computer brains are, and let me focus on things my human brain is good at as well as things I just would rather think about.

Re: 25 Years In Speech Technology and I still don’t talk to my computer

#114

Post by Godel_unicode »

dheera wrote: post #59
dreamer7 wrote: post #41

Earlier quoted context omitted.

That doesn't seem correct. People prefer apps because they are better at providing the information we need (menus, products, services etc) and the alternative for engaging through calls is either talking with overworked humans or a painful IVR service. For me, the ideal interface to engage (order something, create a note etc) when you know what you want is through speech. It just feels so effortless.

I can definitely say that for food I MUCH prefer the web/app interface. I can swipe through a bunch of nice-looking food and just tap on something that looks appetizing. You can't get that with voice no matter how you do it. It's like if a restaurant had no menu and the wait staff just recited the menu in your face and asked you to pick. Even with real humans, voice isn't the best interface for presenting food choice…

> if a restaurant had no menu and the wait staff just recited the menu

That's exactly how food is presented to you at lots of high-end restaurants. Once you get above the fast-casual tier, menus never have pictures. If you know a lot about food and you trust the chef, voice is a perfect medium for explaining the choices. Pictures are just a discovery aid.

Re: 25 Years In Speech Technology and I still don’t talk to my computer

#115

Post by tjungblut »

mikestew wrote: post #73
dheera wrote: post #5

A big problem with the assistants is that as soon as they fail at a query they seem stupid, I feel stupid, and I stop using them for a long time. Last weekend I had the following failed queries: "OK Google, what is the air quality like at Mt. Shasta today?" "OK Google, add a waypoint for the last gas station before the mountain pass" "OK Google, what percentage of people can you detect to be wearing masks on recent I…

A big problem with the assistants is that as soon as they fail at a query they seem stupid, I feel stupid, and I stop using them for a long time. Raise your hand if you remember the early days of Xbox "recognition", as you listen to other people in your party: "Xbox, record. XBOX, RECORD. X...BOX, RE...CORD." It's the reason that "oh, we can't sell an Xbox without a Kinect" Kinect went into the garage after a few mon…

What I find even worse is when you play FIFA and some random non-related british commentary triggers the "xbox" speech command. Infuriating, especially since you can't turn it off.

Re: 25 Years In Speech Technology and I still don’t talk to my computer

#116

Post by h2odragon »

smnrchrds wrote: post #92
bluedino wrote: post #80

I'd like to go back in time and fire up Dragon Dictation and see how far we've came. The Pentium processor's power over the 486 was supposed to be the missing link to 'working' text to speech.

Please let me know of the results when you do. It unfortunately doesn't have a free trial.

Petty gouda wince thestraining fun.

AFAIK it was at the level of super funky chording keyboards; with a lot of investment it could be a cool and impressive input method, but the personal investment side was too much for most users. There was an "assistive technology" user base that i dont think they ever realized they had.

Re: 25 Years In Speech Technology and I still don’t talk to my computer

#117

Post by kebman »

I like taking dictates for memos or writing text messages, especially when I'm walking around town and need to pay attention to traffic. It looks like I'm just on my phone as normal, so nobody sees that I'm actually just talking to myself. Sometimes the mistakes the AI maks are hillarios, but usually they're just annoying, as I need to manually edit them. But in the end, doing that takes a shorter amount of time than thumbing it in.

Re: 25 Years In Speech Technology and I still don’t talk to my computer

#118

Post by aloknnikhil »

It's also the unprompted responses that has started to bug me. I don't know if Google is having a bad rollout or this is deliberate but my Google Home is being triggered a lot more often now and I don't recall anything remotely close to the wake phrase being said. Also, the responses to questions when actually prompted are not at all what I'm expecting. Particularly, Google has trouble understanding a lot of context around Spotify playlists. It still cannot distinguish between a song and an album with the same name despite me prepending the request with "Play the Song...". Overall it's just a terrible experience and that makes the whole assistant thing less like an assistant and more like a toddler that refuses to cooperate.

Re: 25 Years In Speech Technology and I still don’t talk to my computer

#119

Post by dguo »

Audio feedback doesn’t give the user the same reassuring sense of certainty as a graphical user interface. One glance will confirm that I have typed my card number correctly, but you don’t have to be unusually impatient for your heart to sink, when you hear the inhumanly calm words, “I heard 4659 1234 1234 1234. Is that correct? Say yes or press one to confirm”.

This is the main reason I rarely use voice controls. Even if voice is faster than typing for most cases, typing never fails catastrophically like voice does.

I could use voice only when I'm confident it will work, but then there's the mental load of making that prediction. It's easier to just always go with typing.

Re: 25 Years In Speech Technology and I still don’t talk to my computer

#120

Post by thewarrior »

iandanforth wrote: post #99
dheera wrote: post #5

A big problem with the assistants is that as soon as they fail at a query they seem stupid, I feel stupid, and I stop using them for a long time. Last weekend I had the following failed queries: "OK Google, what is the air quality like at Mt. Shasta today?" "OK Google, add a waypoint for the last gas station before the mountain pass" "OK Google, what percentage of people can you detect to be wearing masks on recent I…

Because this is HN, I want to say that I think the above is now doable and more people should be working on startups in this space. The thing we're discovering is how to marry NLP (which works pretty well) with structured databases and automated tools (which work really really well). A pure AI play probably won't get this done. But if your AI starts knowing how to use information tools then I think we'll see a lot of…

Sounds fascinating. Could you share any pointers to papers , academic work on this subject ?
Post reply on HN