Week 11: Voice
Most of us are aware that some form of voice recognition software exists. However, not a lot of us have actually tried it, so this section is designed to get you to interact with some form of voice recognition software during a regular daily activity, and then record your experiences, (both positive and negative), with the software. Using Vocaroo, record your experience and post the link that Vocaroo gives you below.
Posted in: General, Week 11:
manny 12:21 pm on November 13, 2012 Permalink | Log in to Reply
Hi all,
I decided to briefly discuss some of the pros and cons to using the Dragon Dictation app. The Vocaroo software asks for access to the camera too but I don’t believe it makes a recording, just wondering if anyone else experienced that. Anyways, enjoy……
http://vocaroo.com/i/s06Hl7yqHMow
joeltremblay 8:36 am on November 14, 2012 Permalink | Log in to Reply
Hi Manny,
I actually use the PC software to do essays and film reviews. It definitely works better than the app because it has the chance to learn from your voice and the accents therein. I’m curious if you think it changes accuracy due to accents etc. because I’ve found the Iphone app to be less accurate when I drop into Canadian drawl.
rebeccaharrison 11:09 pm on November 14, 2012 Permalink | Log in to Reply
I believe my hyperspeed drawl is also unappreciated. Here’s my take on it. I could see it being really fun to do with friends. For instance: who can come up with the most wildly inaccurate dictation. I’m really interested in what it does for education in the future though….
http://vocaroo.com/i/s1oVYqB9M5Uo
Mike Rae 3:58 am on November 15, 2012 Permalink | Log in to Reply
I had a similar experience Rebecca. Not perfect, and I found myself trying to do a recording that was perfect. My school is all ESL, so I doubt too that I would be able to use it in my teaching right now.
joeltremblay 9:06 am on November 15, 2012 Permalink | Log in to Reply
Interesting that it came up with curse words Rebecca and Mike, especially since it pulls from your phones current dictionary (The one that you constantly update when you send texts etc.) 😀 Sorry it wasn’t more useful for you. Do you think the tone of voice rather than the speed had anything to do with it?
visramn 4:47 pm on November 13, 2012 Permalink | Log in to Reply
Hi,
I used the app Vling. Click on the link to hear my views on this app. http://vocaroo.com/i/s1qXtWcNco2p.
.
Thanks,
Nureen
joeltremblay 8:40 am on November 14, 2012 Permalink | Log in to Reply
Hi Nureen,
I’ve found that Vlingo is definitely accurate and is quite useful for hands free control of your phone. What were you using it for specifically? (feel free to reply via vocaroo if that’s easier for you).
visramn 8:06 pm on November 15, 2012 Permalink | Log in to Reply
I used it for emails, texts, to navigate on the internet, etc. I found that it could not understand names and kept pulling up different names. But that could be because many of the names in my address book might not be in the word bank they use.
Nureen
Jenny Brown 6:00 pm on November 13, 2012 Permalink | Log in to Reply
I uploaded Vlingo to my iPhone and found it easy to use but not accurate. For more of my thoughts click on the link: http://vocaroo.com/i/s1Cj45Ri7g8C
joeltremblay 8:42 am on November 14, 2012 Permalink | Log in to Reply
Hi Jenny,
I use Dragon on my Iphone over Vlingo usually although I have both. I’ve found that the accuracy is a little bit better as far as sentence structure goes but it doesn’t have the hands free functionality that Vlingo does unfortunately. Thanks for trying it!
teacherben 7:19 pm on November 13, 2012 Permalink | Log in to Reply
I recently had a pretty good experience using gesture and voice with my android phone. There is a keyboard called Swype that everyone should try. It is fantastic. Instead of typing on individual keys, you put your finger down on the first key of the word, then slide to the next letter and the next without lifting your finger. After just a few minutes of playing with it, I was typing about 20 words a minute on my cell phone. I have now been using it for a couple of years, but it was recently acquired by Nuance, the same company that makes Dragon, Naturally Speaking software. Between the two options, I was able to take an idea that I had in my head for a picture book, and get the whole thing written, page by page, with just a cell phone.
One thing I did find however is that different writing mediums seem to promote a very different flow of ideas. The research seems to bear this out. For example, here’s an article that discusses how handwriting seems to stimulate the brain much more than typing:
http://www.nhs.uk/news/2011/01January/Pages/writing-versus-typing-for-learning.aspx
As someone who once aspired to be a writer, I definitely found that stuff I typed and stuff I hand wrote was very different in style. Now, with this sort of sliding, gesture-based keyboard and voice dictation on the table, it will be interesting to see what new research reveals.
joeltremblay 8:35 am on November 14, 2012 Permalink | Log in to Reply
Hi Ben,
Isn’t that similar to what Samsung has done recently with their smartphones? Interesting that they have tried to apply to keyboards as well.
teacherben 4:37 am on November 15, 2012 Permalink | Log in to Reply
They just included the Swype keyboard by default instead of the stock Android one. Same keyboard.
Jonathan 9:53 pm on November 13, 2012 Permalink | Log in to Reply
Quick review on Siri (Iphone), Siri (Mac) and Google (iPhone). I don’t usually use voice commands through my phone so it was nice to have an opportunity to play around with it again. I’ve created a Vocaroo (thanks for introducing us to it — super neat) about my thoughts:
http://vocaroo.com/i/s01JsQakIitT
The surprise find for me was the Google one in the end. I found it was able to predict my speech fairly accurately and quickly. I did a bit more research and found that indeed it was fast. Take a look at this one video where the person races Siri against Google!
http://gizmodo.com/5956433/google-voice-search-vs-siri-whos-the-best-iphone-assistant
joeltremblay 8:44 am on November 14, 2012 Permalink | Log in to Reply
Hi Jonathan,
One of the benefits of any google software tends to be the accessibility of it and how much more the developers listen to the community so I’m really not that surprised by the gizmodo article although I wasn’t aware of that in the first place. Thanks for thinking outside of the box and using Google voice as I completely forgot to include it 🙂
Scott 6:51 pm on November 14, 2012 Permalink | Log in to Reply
I concur with Jonathan on this one, internet searches on my iPhone 4 are a pleasure with the most recent voice enable search app from Google. The speed in particular is simply remarkable.
While I have found Dragon Dictation and Search convenient at times in the past, the lack of full integration with iOS, limits its overall usefulness I think. Which highlights an annoying problem many people have recognized, with technology of all sorts today, namely that because tech giants, such as Google and Apple, have ‘issues’ with each other on a bushiness or IP level, we the customers often suffer as as result.
Google’s voice technology appears to be superior to Apple’s at present, or at least for internet searching, yet Apple’s on going patent disputes with Google and other manufactures, mean that in some cases Apple customers are being forced to use an inferior product. This very situation recently played out with the whole map fiasco in iOS 6, which has left Apple customers with a poorly implemented map app compared to Google. Similarly, I believe my iPhone 4 has the hardware capability to use Siri, however software licensing constraints with Nuance, prevent the phone from being able to use this innovative new feature.
As a form of NUI, voice control is very exciting and still in its infancy really. I only hope that there can be more ‘democratic’ implementations of the technology in future.
BTW, I wrote this entire comment quite efficiently using Apple’s built-in voice dictation function, one of the stand out features in its latest desktop operating system, Mountain Lion.
Happy dictating all.
joeltremblay 9:07 am on November 15, 2012 Permalink | Log in to Reply
Interesting the complexity of the speech above. Especially abbreviations I’ve found that voice recognition software has big issues with. I’m curious if you were doing a lot of editing or was it just really accurate?
kstackhouse 10:42 am on November 15, 2012 Permalink | Log in to Reply
Great points Scott. I have not upgraded to Mountain Lion yet. When I was researching the Lion speech commands info it seemed pretty basic. It reminded me of those functions available to my iPhone 4. I think for many of these features time and resources will only make this technology better. As you mentioned they are still young.
joeltremblay 8:56 am on November 16, 2012 Permalink | Log in to Reply
I’m interested in just how different the functionality of the two operating systems are. I wonder if there are net comparisons out there as all I was able to find were comparisons between dragon and Mountain Lion or someone talking about the benefits of it:
http://www.gottabemobile.com/2012/07/29/osx-mountain-lion-review-dictation/
avninder 3:54 pm on November 14, 2012 Permalink | Log in to Reply
My vlingo review
http://vocaroo.com/i/s1IAQzWBrzEU
joeltremblay 9:13 am on November 15, 2012 Permalink | Log in to Reply
How was Vlingo as far as commands etc.? What worked for you and what didn’t?
avninder 4:15 pm on November 14, 2012 Permalink | Log in to Reply
A humorous clip showing how voice activation is far from glitch free:
http://www.complex.com/tech/2011/10/video-jack-donaghy-predicts-the-future-of-television-sets
Mike Rae 3:47 am on November 15, 2012 Permalink | Log in to Reply
love it…Jack is great, and I agree with him, not a big fan of the remote getting lost. I always thought that a page button like on a cordless phone would be nice if it was built in.
sophiabb 11:25 pm on November 14, 2012 Permalink | Log in to Reply
My Vocaroo review
http://vocaroo.com/i/s0WuSHJyHCu1
I tested this voice software for podcast possibilities. I found it to be very clear. I liked that I was able to download as an MP3 and was able to open in Media Player as well as Quick Time. I was also able to edit it.
Sophia
joeltremblay 9:10 am on November 15, 2012 Permalink | Log in to Reply
HAHAHAHAHAHAHAHA… I was trying to figure out exactly what the hell this was!!! Too funny 😀 I’m guessing vocaroo was useful for you if you’re using for something other than this course 🙂
sophiabb 11:39 pm on November 14, 2012 Permalink | Log in to Reply
Here is the correct link: http://vocaroo.com/i/s11GVVOnlOM6
Sophia
joeltremblay 9:12 am on November 15, 2012 Permalink | Log in to Reply
I’ve actually started using vocaroo for a lot of activities in the class including film reviews etc. Something else you can do is when you go to the link you right click on the player and save as an MP3 as well if you accidentally click the wrong thing.
sophiabb 12:40 pm on November 15, 2012 Permalink | Log in to Reply
Thanks for the tip.
Pat A Son 11:55 pm on November 14, 2012 Permalink | Log in to Reply
My Voice talk review
http://vocaroo.com/i/s1hT1k5xVdlt
joeltremblay 9:14 am on November 15, 2012 Permalink | Log in to Reply
It’s a little strange talking to the computer sometimes. My students always look at me funny when i’m doing it.
Pat A Son 11:51 pm on November 15, 2012 Permalink | Log in to Reply
That’s because the technology is new to us and we haven’t gotten used to it yet. Most people I am sure would prefer to control the computer by talking to it that typing or using a mouse because for us talking comes more naturally. Furthermore with the push by some software maker to have the user personalize their software at the interface level in a way makes the user experience a personal one. This fits in with the idea of personal learning that we looked at here before and all the option that we looked at this week voice has the potential to make computer usage almost like communicating with another person. Maybe this can make the difference for learners who are technologically challenged in one way or another.
As I said on vocaroo the current offerings are all dead and robotic when the day comes that CG generated voice sounds more natural and is more customizable it serve as the pathway by which a learner may be able to grasp a difficult concept.
Mike Rae 3:39 am on November 15, 2012 Permalink | Log in to Reply
dragon dictation review:
http://vocaroo.com/i/s1BhC2TP7PjD
joeltremblay 9:16 am on November 15, 2012 Permalink | Log in to Reply
Thanks for the participation Mike,
Interesting that you had some problems with the exclamation and the curse word censorship. My wife and I tend to use it for shopping lists etc. and have ceased using pen and paper and instead just speaking at Dragon and texting the list to each other.
jhodi 7:46 pm on November 16, 2012 Permalink | Log in to Reply
Hi Mike,
I quite enjoyed listening to your experience. I also tried to get punctuation and I found that some things worked such as saying ‘period’ or ‘exclamation mark’, but that does make me wonder what if I meant period in the sense of a hockey game or something and not the punctuation? I found some issues with the app’s accuracy. I was frustrated with the app in that I would have to read over what I said and make corrections to the words because it would have totally different words than what I said. I do wonder how this app would work with different accents?
Jhodi
kstackhouse 10:38 am on November 15, 2012 Permalink | Log in to Reply
Here is my recorded reflection on using Siri. http://vocaroo.com/i/s0WQybGWkDuy
Thanks for providing this activity. It gave me an excuse to spend some time trying to use Siri, which I have not done before.
joeltremblay 9:00 am on November 16, 2012 Permalink | Log in to Reply
The issue about quick responses seems to be wide standing which is really interesting considering that apple always pushes the response time during their commercials. False advertising maybe?
kstackhouse 10:48 pm on November 16, 2012 Permalink | Log in to Reply
I’m not sure if it is false advertising or optimistic that one will have a good signal all the time. When it comes to simple things, like playing a song or dialing a number the response time was fine. It was when I asked for more detailed things like the movie listings or directions that there seemed to be more pauses. I think we have gotten spoiled with the immediacy of Google, Bing, and Yahoo searches that if something takes 4 seconds we start to think it is broken.
This reminds me of comedian Louis C.K.’s talk when on the Conan O’Brien Show on technology and how it is wasted on us. The bit was called “Everything is Amazing and Nobody is Happy”. It seems as though the original has been removed from youtube, but you can watch his stand-up version but it is not censored.
He also does a great bit on Twitter, social networking, and smart phones. He talks about how we are living our lives through the little screen on the camera. This is similar to the point I was making about the AR goggles in another thread.
jenbarker 11:14 am on November 15, 2012 Permalink | Log in to Reply
After reading Scott’s post I was really excited to try dictation on my iMac but unfortunately although I have upgraded to Mountain Lion, it doesn’t appear that I have the dictation feature. In my system preferences I only have “Speech” as an option. Scott, I am surprised that your text was written with correct punctuation. Did you have to edit it?
I am going to try Siri on my husband’s iphone tonight and will post my Vocaroo later. In the meantime, what is the name of the Google app everyone seemed to like? Do you download it from the app store and does it cost anything?
teacherben 6:41 pm on November 15, 2012 Permalink | Log in to Reply
You enable voice commands in your Mac in the Accessibility section in the system preferences. It’s the last item on the list–“Speakable Items”. (I am on Mountain Lion too–btw, have you noticed a loss of battery life since you upgraded? My Mac was only a year old and my batter went from 6 or 7 hours down to about 3 after the upgrade, and it’s sluggish and awful. i wish I stuck with Snow Leopard.)
jenbarker 7:46 pm on November 15, 2012 Permalink | Log in to Reply
Yes Ben I have noticed the same thing. I don’t go anywhere without my charger. BTW I don’t have an Accessibility section. I went to system preferences, and under the section system, I have a microphone labelled Speech. I clicked this and was able to turn speakable items on and off but it won’t allow me to convert voice to text. It only answers a couple of questions.
teacherben 7:50 pm on November 15, 2012 Permalink | Log in to Reply
In my system preferences, in the system section, I have an icon for Accessibility and I also have an icon for Dictation and Speech. Once enabled, I just have to click on fn twice to turn it on.
adi 6:42 pm on November 15, 2012 Permalink | Log in to Reply
Hi,
As a language teacher I have always been interested to see what voice recognition machines do when you speak with an accent. I downloaded Assistant Personal Secretary for my phone and said: “Hello mate, got a light? Oh, and could I have a glass of water, please?”, but it clearly did not understand my ‘cockney accent’ (http://vocaroo.com/i/s0rCj0CL38QM). This voice recognition programs that transcribe what they hear still have a long way to go, particularly if you take into consideration that there are more speakers of English as a Second Language than native speakers. Still, it was a lot of fun. Vocaroo, on the otehr hand, I have used in my classes and think it’s a great tool.
teacherben 6:48 pm on November 15, 2012 Permalink | Log in to Reply
As the technology gets cheaper and more accessible, I think we will see this change. While it is possible to ‘train’ these programs to better recognize your voice, we are likely to start seeing add-on packs and so on for more regional accents in the same way that we already do for different languages. Many of these programs are already modular and allow users to add additional functionality by downloading extra bits.
teacherben 7:39 pm on November 15, 2012 Permalink | Log in to Reply
Here is an interesting overview of accents research and voice recognition software. It was an interesting read and outlines some of the challenges in terms of getting it to work:
http://www.phon.ucl.ac.uk/home/mark/accent/
As he explains, an accent mismatch between the speakers used to ‘train’ the software as it was developed and the user can lead to up to a 30% error rate. And it would appear that the solution is not going to be a one-size-fits-all program but rather by leveraging the modular nature of the software and including ‘language packs’ that may be based on specific accents rather than different languages.
I am willing to bet that once some open source players get in on the game, then individual groups will be able to create their own language packs based on a very specific regional accent. For example, as this technology is integrated into the browser, as happens with most technologies sooner or later, a company like Mozilla or Google may use an open architecture that will allow these sorts of plugins.
joeltremblay 9:55 am on November 16, 2012 Permalink | Log in to Reply
Hahaha… I like the translation that it provided for you. I wonder if the accuracy is representative of the entire population or if it is just localized to the Cockney accent? Also, Dragon for your home computer actually learns your accent and builds up a dictionary the more you use it so as you use it, it trains itself to your voice.
Lisa Nevoral 8:08 pm on November 15, 2012 Permalink | Log in to Reply
Hey All,
Here is my Vocaroo: http://vocaroo.com/i/s0Tc6DOwnwOj. Note-I have a cold and did not use a microphone, so quality not so good.
I have recently purchased a Galaxy S3 and of course there is voice recognition program with the phone. I have two options – one that is like Siri in Apple (S Voice or Galaxy) or a Google Option. The S Voice gives me options to use voice recognition to do things with the phone, while the Google Option allows me to search things on the Internet. The S Voice is pretty finicky. It doesn’t always do what I want, but maybe I need to pronounce my words better. I can see where voice recognition could come in handy (such as using it for a “To-Do list, texting a message to someone, etc.). I think I just need to play around with it a bit more.
Lisa
joeltremblay 9:58 am on November 16, 2012 Permalink | Log in to Reply
Interesting the difficulties you had with it. I wonder if that’s only on the G3 model since most people have reported google phones being superior as far as voice recog goes with reference to the comparitive apple products?
Lisa Nevoral 7:44 pm on November 16, 2012 Permalink | Log in to Reply
Hey Joel,
S Voice is a personal assistant and knowledge navigator for Samsung Galaxy SIII and Samsung Galaxy Note II. Apparently, it can help with opening apps, setting my alarm, updating Facebook, telling me the temperature, help send texts or place calls. I just find that it doesn’t always recognize the words I say very well. The Google option on my Android seems to understand what I’m saying much better and is quite fast finding me information.
Lisa
teacherben 5:20 am on November 16, 2012 Permalink | Log in to Reply
A company called Maluuba, that makes a voice recognition app for Android that is apparently better than Siri, just released their API so that other apps can tap into its functionality. So if you were working on an app, say an ebook, you could make it so that if the user has maluuba installed and running, they could interact with the book, turn the pages, whatever, using voice. Nice.
http://mobilesyrup.com/2012/11/14/maluuba-launches-natural-language-api-to-bring-siri-type-assistance-to-any-app/
C. Ranson 3:31 pm on November 16, 2012 Permalink | Log in to Reply
I downloaded both Vliingo and Dragon Dictation to my iphone through the itune store, took seconds. I had difficulty accessing my contacts through Vlingo, the voice recorded messaging worked well. I then downloaded Dragon Dictation and this application works fabulous. The voice recording of message is very accurate, with each pause a comma was added and you had to indicate a period. http://vocaroo.com/i/s0ZD6FuHKvrI
Catherine
joeltremblay 2:09 pm on November 17, 2012 Permalink | Log in to Reply
Hi there Catherine,
Good to hear that you spent the time to actually compare the two services and that you were able to ascertain which was more compatible with your understanding and abilities.
jhodi 7:40 pm on November 16, 2012 Permalink | Log in to Reply
I downloaded the Dragon Dictation app on my iPhone and found some positives and negatives to the app. I also noticed afterwards that this same program is on my work computer (I had seen the logo before, but had never tried opening it), and tried it on there as well. Here is the link to my Vocaroo recording:
http://vocaroo.com/i/s1w5koO8gXIP
Jhodi
joeltremblay 2:08 pm on November 17, 2012 Permalink | Log in to Reply
Thanks for the reply Jhodi. Good to hear that you attempted a couple different options and didn’t get frustrated and walk away from it the way I might have done in your situation.
jenbarker 11:48 am on November 17, 2012 Permalink | Log in to Reply
I tried to use Vocaroo this morning but after ten times I gave up. I am not sure if it was due to a setting I had on my computers (I tried both my laptop and iMac) but my voice sounded mottled and was almost inaudible. So instead I used a screencast program called Jing that I love. It records a portion of your screen and allows you to talk to it. I use it when responding to student’s papers and/or lesson plans as you can scroll through the document on the screen and use your cursor/arrow to point things out and talk to them simultaneously.
On my Jing I gave a review of Siri. Here is the link to my Jing. http://screencast.com/t/Alj9CGtd
joeltremblay 2:05 pm on November 17, 2012 Permalink | Log in to Reply
Hi there Jen,
I was having problems with it yesterday and the day before at the school but I believe that was due to the internet connection and more importantly, the uploading ability of your line. It could also be due to Vocaroo itself having issues but who knows. Thanks for adapting and posting the alternative Jing though as I’m sure other people have had this issue.
jameschen 11:29 pm on November 17, 2012 Permalink | Log in to Reply
My Vocaroo recording on my voice recognition software experience is at http://vocaroo.com/i/s0zSG8gFEUqy.
I think voice recognition technology would benefit students with a disability who are unable to write/type using their hands. Another educational benefit could be to help English Language Learners improve their spoken English by trying to train the software to recognize their voice. Other than this I wouldn’t recommend educators to make use of the technology – unless they have extraordinary patience.
James
joeltremblay 1:09 pm on November 18, 2012 Permalink | Log in to Reply
Hi James,
Because of the quality of the vocaroo recording, you have a somewhat Siri esque quality to your Vocaroo recording. It’s interesting as it sounds like you typed it and a software speak it for you. Sorry it didn’t work out well for you, but we appreciate the perseverance.
joeltremblay 1:11 pm on November 18, 2012 Permalink | Log in to Reply
Alright everyone,
As your guide through the mystical lands of futuristic and sometimes prehistoric voice command software, I bid you adeau and thank you for participating in our activities. Thanks you for your insight, ideas and professionalism. Cheers!
melissaayers 2:31 pm on November 18, 2012 Permalink | Log in to Reply
Thanks to the group for introducing this activity, as well as introducing us to the Vocaroo tool.
I have been interested in voice recognition for many years (since I started my computer science degree in fact) but have been disappointed by how slowly it had evolved over the years. However it look likes significant improvements have been made thanks to the introduction of competing mobile devices I think.
I tried Siri a while ago but like many others have reported here the experience was not too fantastic and I gave up not long after I started. I found it made too many errors.
In contrast I have recently been playing around with the Google Voice and find it amazing, the advancements in this technology are clear and I am sure the improvements will keep coming for both this product as well as in Siri and Samsung’s version. It’s something I will definitely consider using in both in a classroom and daily life.
Just for fun I tried Dragon Diction with the language set as French (Canadian) and it was a complete disaster, I know my accent in French is not that great but it was really nothing like what I said – not even a word. Once I switched it back to English it was a lot better but still made quite a few mistakes. I think I would need to get used to it a bit before I find it completely useful.
Overall I found that google voice was the best product for my dictating (and searching) needs.
joeltremblay 11:42 am on December 7, 2012 Permalink | Log in to Reply
The best part about the desktop version of Dragon Dictation is that it keeps track of the different nuances and tendencies that you have within your diction. The mobile app is somewhat limited in that capacity unfortunately and because of that you get the mistakes you talk about. Just last night I was using it for something, but was quite tired because of a particularly gruelling schedule of late and it was making all kinds of mistakes.