Technology to turn text into speech already exists. But what about the way around?
A new product just caught my eye. For hearing people, using SpinVox would be like having a personal secretary transcribing your speech. It can convert spoken words into a blog, e-mail and the such.
It means that the technology to instantly caption everything from obscure DVDs to podcasts to CNN.com video clips is here.
Should it migrate to portable devices, its ramifications for deaf people are great. We’ll be giving directions to lost hearing people in no time.
© Copyrighted material. This article cannot be copied, reproduced or redistributed without the express written consent of the author. As with every blog on this website, this blog does not reflect the opinion of DeafDC.com.
17 Comments
Sorry, the comment form is closed at this time.
Yep. Like I said, technology is key which will allow deaf/hh people obtain greater independence. Not to mention greater enjoyment when it comes to entertainment, too. The “on demand” speech-to-text will only get better because of faster and more efficient computer chips design. These would be the new “core” computer chips that can generate “Millions of Instructions per Second” (MIPS). The goal now for core chip manufacturers is get 1000s of cores on a single computer chip. This would be a natural evolution on design efficiency and performance that will use less electricity while getting more computational power at the same time.
Presently, quad core computer chip is next. The next leap would be to design a multi-core processor where it would get us into the Tera-scale computing getting up into 10 TIPS by 2015 which is Intel’s goal.
So, tell me, what will computing power and performance be like in 5 to 7 years (2012 - 2014)? Who would’ve predicted a miniature version of the MPP on a chip 10 years ago that would be the key on increasing MIPS to TIPS (Trillion Instructions per Second)?
In short amid all this mumbo jumbo would be a vastly improve computer performance. This means very good news for us in the very near future.
Hi,
This doesn’t seem too exciting to me. It’s a website offering voice recognition via telephone and wiring it up to blogs, emails and so on.
Would have to see how well their technology would work with DVDs and podcasts and so on as those kind of things may have multiple people talking at once and would be much harder to transcribe properly.
SpinVox seems like a cool service but I don’t thinks it’s ground-breaking by any means.
i’m not talking abt now. my point is that this technology has capacity to transcribe speech and has potential for more widespread usage in the future. i’ll need more info if it can transcribe that well right now without intensive “training”- more comments from those who have used it will be appreciated.
btw, it already does podcasts, but it costs money. 75 cents a minute. it also automatically transcribes podcasts at livejournal if they’re less than 3 min long (http://www.stonedeafpilots.com/?p=58).
I don’t know. The way SpinVox as it stands now is still not quite practical for those of us who are deaf or hard-of-hearing. Still, it will be interesting to see how this all evolves.
My sister who is hearing uses “spinVox” technology; as it stands now, you must “teach” the software how to “identify” your voice. It would be very difficult NOW to have instant captioning but who knows what the future will bring.
My sister is able to write a whole newspaper (on line)via this software.
see for yourself at http://www.hullmagazine.com
human
really? i didn’t get that impression when looking at spinvox’s website. see that link with which u’d be able to immediately test the system- u call in to utter something and your comment will appear on the screen on the site? maybe when u do the test, u go thru a quick training session- i wouldn’t know.
disclaimer: i havent checked out the spinvox website…
voice-to-text software has been around for a while. the most popular is DRAGON (IBM), and i’m unsure if that’s an acronym or not. There’s another brand but DRAGON has like 80% of the market. In any case, both brands require time with the software so it “learns” your speech patterns. It’s also stand-alone so if I were to borrow your laptop and it had DRAGON installed, I would be a stranger to your licensed copy.
This is probably because speech discrimination by AI isn’t anywhere close to the reception/performance we can expect from any human. Not surprising since we would expect the same from an automated ASL interpreter, you know?
One immediate application for this is self-speech training. If people with perfect speech aren’t understood by the software right away then I assume you have to take great pains in order to be understood. It follows that when we’ve come to the point where it understands words, sentences, and even entire passages, then either anyone off the street should understand us… either that, or nobody will since accommodating the software could lead us to speak with a convoluted accent all our own?
heh. “a convulted accent all our own”. but software that specializes in speech training may have some potential. no more pressing ur hand on a vibrating triple chin.
i’ve also been aware of DRAGON. what may be new about spinvox is that it doesn’t seem to require time to learn one’s speech patterns. one commenter says it does. maybe he/she’s right. but the fact that it automatically transcribes podcasts @ livejournal definitely gives me the impression that it doesn’t. like i said, i’d appreciate more info.
and do check out the screen on the website. of course, for all we know it’s really human transcribers doing the work, a la malzel’s machine.
Went to their website. Yeah, seems like no prior interaction with the software is necessary.
When I do find a mic, I’ll take their “test drive.” And for my first attempt, I’ll be sure to intone “With all due respect…”
Do you notice how we’re not quite so “insular” when it comes to technology that doesn’t develop from the assumption that deafness is the problem? Suppose cochlear implant technology develops at the same rate as this Spinvox thing or something like it develops over the next ten years. Is a deaf person who refuses an implant but eagerly embraces Spinvox technology still “insular”? Still living in a little Deaf/ASL bubble far removed from reality?
We’re a lot more dependent on technology in order to be that more independent ourselves. Doesn’t matter which technology you use, if you’re going to be independent, you’ll be needing them. Please note, I said “technology” which can be anything and everything.
Irony at work! In order to become more independent, we must become more dependent!
Hahaha!
Yep. I depend on my hearing aid for everyday communication with hearing people just as people depend on their Sidekick or Blackberry, relay operators or an UbiDuo to communicate.
Yes I agree with a lot of your comments above. You all have brought up good points and views toward technology. I on the other hand like some said we depend on technology to help us such as pagers (SK or Blackberry and so on and VP, AIM and AIM relay) and so on… but I always fear technology is only to increase to make life easier for the hearing working community while it does help us but when we use it applied to the hearing community, they think we lack something they have benefits of.
Everytime they come up with newer technology, I am behind one step which keeps me from getting a job. I was denied to work on films due to the fact that all assistants must wear a portable headsets to walk around the sets, if this technology did not insist, I’d have this job working on a major film for Warner Bros. and the fact that everytime we come up with a new electronic toy, it is so expensive, for struggling people like me take years to get it when the price goes down, as we do get it, there is a new gadget out and the gadget I waited years to get is now old-fashioned! For example I wanted to get sidekick and couldnt’ get it and 3 years later I finally was able to get it, SK3 is out. I cannot win with technology especially if $$ is an issue or if it is used for a job. At the same time, using my pager definitely helped me work in the office at the studios a lot easier which was a relief to me and the staff but for me to advance on the set, I must use headsets or I was repeatedly told why I cannot be hired is cuz “this is a fast paced set, which is why headsets are required” and the director mentioned fast-paced repeatedly like I’m slow-minded. Oh well, hopefully a technology will come out someday where it’s a walking interpreter. LOL.
My Uncle Ross (Gally Alum) was one of the first people who focused on developing and applying this technology for the Deaf. (Actually, for the hearings that didn’t know sign. ;) )
http://www.netac.rit.edu/publication/asr.html
From RIT’s website:
>>In the absence of practical automatic speech recognition systems, NTID has been active in the development and application of operator-assisted speech recognition since 1981 when it acquired the first model of a stenographer-assisted speech-to-text system. ASR was introduced in selected NTID/RIT classes in February, 1982, and with large audiences of deaf adults at a convention of the A.G. Bell Association for the Deaf in Portland, Oregon, in June, 1984.
While there is no question that technology has come a long way and played a vital role in reducing many of the barriers that exist for the Deaf Community, I still feel uncomfortable with those who would chant “technology is the answer” as their personal mantra.
Yes, technology will help to enhance our sense of independence. But if I might refer back to Shane Feldman’s blog about hearing aids and earning capacity, it is as some of the comments to that post have mentioned - as great as such technology might be, it cannot resolve the attitudinal barriers that we still encounter in the hearing world.
While this technology might indeed reduce those communication barriers that are the primary thing that most hearing people are concerned with, they will not necessarily raise the level of sensitivity, understanding, and acceptance of deafness, Deaf Culture, or the Deaf Community.
This can be clearly seen in the example shown on yet another DeafDC post - Bobby Cox talking about the use of IP-Relay to make prank calls.
While I am all for on-going research into using technology to benefit the Deaf Community, I also think it has to be balanced - as Shane says - with educational programs and ongoing efforts to reduce those attitudes and biases which serve as the actual barriers.
Persons who hang up on relay calls, companies that refuse to caption their videos, employers who won’t purchase ttys or assistive listening devices…
All the best technology in the world won’t help us if we can’t resolve problems such as these.
Spinvox… is all about convenience. Who would want to type for themselves unless they absolutely had to? How about getting your proposals or reports typed up in just a minute without having to worry about spell check? This technology is going in an opposite direction of what we would like to use as a tool in interacting with hearing people. Correct me if I’m wrong, the only technology that converts text to voice right now usually are for automated messages generated by something like Interactive Voice Responses (IVR) or text-to-voice messaging. People are expected to speak for themselves, and people who depend on something to speak will probably always be seen as the weird ones.