Speech recognition, anyone?

   Last week a developmental reading instructor popped down to my office bemoaning the fact that "none of my students can type!"   We talked about free online typing improvement sites but also speech recognition, and a quick perusal netted me this from the Quality Indicators of Assistive Technology https://pub.lucidpress.com/2f091482-82da-44a9-b8da-0f9c52f81482/#kDXU0y0swYzB   It's in some weird online format but you can download it as a PDF.   

     I've talked to another staff person and we're going to try the training out on each other first ;)   ... I've already discerned that with my laptop, the built in speech recognition in Microsoft works great in Microsoft Word but doesn't work *at all* on my Wordpress blog -- and does not discern the word "Wordpress" even on word.   My cynical side gets activated but ... I'm home today and will try tomorrow w/ a good headset. 

    I especially appreciate that the training is targeted at students who may not decode well, may not speak clearly, and may struggle with language generation and sentence formation and organizing ideas.   The idea of doing things one sentence or phrase at a time:   "Think it... Say it.... Check it... Fix it..."  seems powerful to me.   

    Their sample sentences don't work for me ... and remind me of years ago when I worked w/ students on language generation ("can you finish this sentence with 15 different phrases that answer the 'where' question?"  I found a quarter _______________)..... 

    Has anybody gotten students using speech recognition for academics?

Comments

If you learners are comfortable with the google tools, the voice to text option in each tool is very good. Not only does it pick up the Down East Mumble I have in my neck of the woods, it also picks up correct words from those with a Hispanic accent. Even better, if the person uses the same machine over time, Google's AI starts adapting as corrections are made in the text. Sorry to be always beating the Google drum, but the voice to text has removed so many typing issues for students and teachers I work with. 

Speech recognition can be a useful tool for students who have difficulty with typing - or other aspects of transcription like spelling. We published a study of speech recognition with high school students with LD in 2004. Back then speech recognition was just becoming powerful enough for practical application. It took about 30 minutes to train Dragon Dictate to understand your voice. And students had to learn some dictation techniques. But the students with LD wrote significantly better essays using speech recognition than using regular word processing. It made no difference for non-LD students, so we argued that it was a legitimate test accommodation. The speech recognition on our smart phones and laptops today is much more powerful. 

It is important to recognize that learners need to learn to use speech recognition and adapt their writing process somewhat. For example, it's best to mentally think of the whole sentence, or phrase before dictating because the tool doesn't not work well with hesitant speech. Also, it is important to have a plan worked out for what you are going to write, again because the tool works better if you are ready to dictate without too many pauses and false starts. Of course, planning is a good idea anyway. A practical issue is that it's difficult to use speech rec in a classroom or other public setting because of noise and also because no one wants others to hear their struggles with writing. 

Overall, I would encourage teachers and students to give speech recognition a try. 

 

 

 

 

Speech to text has come a long way, and there are numerous products you can purchase, but there are also some free, high quality apps that learners may want to try out before deciding what, if any of them, works best for their circumstances.

Best free speech to text apps, according to TechRadar

Mike Cruse

Disabilities and Equitable Outcomes Moderator

michaelcruse74@gmail.com

  1. Google Gboard - If you already have an Android mobile device, then if it's not already installed then download Google Keyboard from the Google Play store and you'll have an instant text-to-speech app. Although it's primarily designed as a keyboard for physical input, it also has a speech input option which is directly available. And because all the power of Google's hardware is behind it, it's a powerful and responsive tool.  You can download Google Keyboard here.

  2. Just Press Record - If you want a dedicated dictation app, it’s worth checking out Just Press Record. It’s a mobile audio recorder that comes with features such as one tap recording, transcription and iCloud syncing across devices. The great thing is that it’s aimed at pretty much anyone and is extremely easy to use.  You can download Just Press Record here.

  3. Speechnotes -  A useful touch here is that you don’t need to create an account or anything like that; you just open up the app and press on the microphone icon, and you’re off.   The app is powered by Google voice recognition tech. When you’re recording a note, you can easily dictate punctuation marks through voice commands, or by using the built-in punctuation keyboard.  You can download Speechnotes here.
  4. Windows 10 Speech recognition - If you don’t want to pay for speech recognition software, and you’re running Microsoft’s latest desktop OS, then you might be pleased to hear that Windows 10 actually has some very solid voice recognition abilities built right into the operating system.  Turn on Windows Speech Recognition by heading to the Control Panel (search for it, or right click the Start button and select it), then click on Ease of Access, and you will see the option to ‘start speech recognition’ (you’ll also spot the option to set up a microphone here, if you haven’t already done that).  

Unlike Michael, I do not have an exhaustive list for iPhone - I can only share what I have tested/used:

1) iPhone/iPad keyboard mic: Any time my iPhone keyboard pops up, there is a small mic icon next to the space bar. This allows me to voice enter text into any empty textbox in Safari, Notes, Messages, etc. It has a limit of about a paragraph before it needs to save and you have to turn it on again. I use Notes in particular to quick capture ideas and then email them to myself later if I need to (keeps me from having a zillion draft emails).

2) Speech To Text: Voice to Text: A VERY plain and simple S2T transcriber that does NOT allow me to get caught up in formatting (but I can change the font size for visibility), and can send the text anywhere through any tool my iOs knows about - or convert it to pdf if desired. No bells or whistles, except the creation of folders to organize your 'recordings.' NO IN_APP PURCHASES or ADS!

3) Dragon Anywhere: Dictate Now: I had this for many years and I was never very satisfied with the reliability or accuracy of the app.

>>> And there are SO many more listed in the App store - have your students try them and see what interface/transcription software works best for them!

 

Just a note that I have only been able to use the Google text-to-speech feature via Chrome browser.  The option does not appear when using Google via Edge, Firefox or Safari.

AND - for those using mobile devices with a non-Google/chrome browser option, there are often built in speech to text options as well as fairly good *free* Speech-to-Text apps your students can try out.

(I've been trying out speech to text to assist me with my ADHD "running brain" - to capture ideas and writing snippets as they occur, rather than either forget them or let them totally disrupt what I am currently doing.)

Keyboarding is an essential workplace skill...sometimes. I would suggest that we need to ask what the essential skill is -- keyboarding, or the ability to use the keyboard for another purpose. In my experience, keyboarding itself doesn't have much value; rather the value comes in how keyboarding is used. In my workplace experiences, keyboarding has primarily been used for searching and communicating in writing. Either of which could be done just as effectively with a dictation tool as it is with a keyboard (once the user is proficient with the tool and the tool has been trained or set up properly for the user). If a student's learning benefits, and s/he can produce the written communication and/or execute an effective search with this accommodation in the classroom, it would be reasonable to expect that the student could use a similar accommodation in the workplace for the same tasks. In the world where we're aiming to provide equitable access to education, work, and the world for everyone, a dictation tool might make that world just a little bit more accessible.

To build on Heather's reply, I would point out that "keyboarding" or data entry is really the ability to *retype* something into a computer (from a printed/written document/sticker, audio recording, or mental count). This is a very specific skill needed for only a specific set of jobs.

All other "keyboarding" type activities are about accomplishing tasks on a computer - and the most commonly available interface is a keyboard and text. Alternate methods of accomplishing those tasks (via alternate process or interface), if similarly efficient, *should be* explored by any student who has learning barriers (whether or not they are documented).

> Writing Tasks - the act of conveying ideas via symbols, does not have to involve a keyboard (or pencil)for he most part. This is where Speech-to-text can really help our students.

> Communication Tasks - the act of communicating information to others, does not have to involve symbols at all - audio and video recordings and even images can work just as well (and sometimes better).

> Learning Tasks - the act of incorporating new ideas and skills into your neural network, can be designed as a stepping stone towards an end goal. For ESL learners, Speech to Text can be used as an assist to improve pronunciation, understanding or English word order, response complexity, etc. Speech to text could also be used to assist a student to learn to EDIT written work - a very common task across many jobs/careers.

I always remember that Socrates cautioned against assuming writing (which he loved) was the only true path to knowledge; insisting that it is only through discussion, argument, and reason *about* information and ideas that we create new knowledge.* I use this as a reminder to always ask "What is the real goal?" and keep my mind open to as many new possibilities (for our learners) as possible.

*Reading, writing, and what Plato really thought, Lane Wilkinson, 2010. https://senseandreference.wordpress.com/2010/10/27/reading-writing-and-what-plato-really-thought/

Dureen and Heather raise excellent points about "what is the real goal" for learners?  While this varies from person to person, adult educators are tasked with building foundational skills.  The term 'foundational' is challenging, as it can seem to adjust as quickly as new technologies enter the workforce.  To that end, I want to draw your attention to the report, The New Foundational Skills of the Digital Economy Developing the Professionals of the Future, published by the Business Higher Education Forum and Burning Glass.  While the publication may appear geared towards higher education leaders, I think it's important that adult educators have an awareness of its contents.

To illustrate this point, "At some moment in the future, many of the high levels of skill that currently seem confined to the upper reaches of the digital economy, or to larger, more complex organizations, will become the norm among jobseekers, incumbent employees, and workplaces. The advanced skills of the past become the foundational skills of the future. In 1880, accountants and mathematicians were the “data scientists” of note, manipulating calculations understood by a select few. Today, many workers across multiple professions perform calculations far more complex, and millions of people understand and manipulate data using tools and techniques that would confound even the most skillful of 19th century experts.​"

Jobseekers and incumbent employees need skills from each of the three buckets (1. Human Skills; 2. Digital Building Block Skills; 3. Business Enabler Skills) to be prepared for the digital economy, but few currently claim this mix of skills on their resumes. Not every person will need every skill, but jobseekers and incumbent employees can mix and match skills to become the blended digital professionals required in a broader economy that is increasingly becoming the digital economy. 

The term, blended digital professionals, could suggest a lot of our teaching about typing, speech-to-text, etc. may look as different between two learners, as their goals for their futures.  My blend may not be the same as the next learners.  The important point here seems to be that we need to not only teach these digital skills, but equally as important, instill a mindset of being open to learning new technologies to do what once seemed improbable for the average person to even attempt learning.

Mike Cruse

Disabilities and Equitable Outcomes Moderator

michaelcruse74@gmail.com 

This discussion brings to mind an experience I had last year when I needed to do some banking business for my mother under a Power of Attorney. It was a small bank branch with an open office layout. Because of the POA, I had to provide a great deal of personal information about both myself and my mother, information that would have been of great interest to anyone who wanted to steal our identities or hack into my mother's bank account. As a security measure, the bank clerk had me write down the sensitive information on a piece of paper. After she keyed in the information, she shredded the piece of paper.

Especially when anyone with a decent smartphone can secretly record conversations, there are workplace activities -- an IRS audit, for example -- where speaking sensitive information out loud isn't a good idea. In the workplace, voice vs. keying is context dependent and one really doesn't outweigh the other.

Robert, you have me thinking . . . 

I am a person who likes quiet when I work.  (When I am washing dishes, this is another story entirely.  I need some 80's music then!)  If someone was speaking aloud to their word processor on a consistent basis, I would lose my focus.  Thankfully, I have never had to work in a cube farm where this might be a reality.  In an office setting, one could just close the door and the issue is non-existent. 

One could also say, "Steve, this isn't about your preferences.  It's about equity in the workplace."  I would have to agree with that person.  

Would anyone else be bothered by the consistent noise of someone talking aloud to their word processor?  What solutions might there be?

Steve Schmidt

Moderator, LINCS Reading and Writing COP

schmidtsj@appstate.edu 

Hello everyone, I wanted to pass along a link to a previous LINCS discussion started by David Rosen on this topic, Teaching Adults Keyboarding Skills: What Works? There are a number of suggestions shared by members in this thread. In David's initial post, you can find a link to David's Literacy List which features--among many other useful resources-- a list of tools for teaching keyboarding.

Please tell us what's been working for you in your practice!

Cheers, Susan Finn Miller

Moderator, English Language Acquisition and Teaching & Learning CoPs

 

I am very happy that I joined this group?. Basically, starting from the day I joined this to now, I can see a lot of improvements. Different people across the world. Different thoughts. I hope I will learn more things in future.