Q. If anyone can use speech recognition software to see their words typed on the screen, how come everyone isn’t using it?
A. Because most people can’t get it to work consistently and quit using it!
Hello, my name is Dan Multer and I've been teaching people to use speech recognition technology on their computer to get the results they want since 2001.
I can teach you to use speech recognition successfully on your computer.
What is the purpose of Speech Recognition?
Speech recognition software makes it possible to type hands-free and control a computer by voice. For people who have limited amounts of time to produce significant amounts of writing, speech recognition software is a power tool that saves a lot of time and energy. For people who can't use their hands or experience pain from using their hands, speech recognition software makes it possible to communicate and connect to the rest of the world, providing significant, life-changing benefits.
Once you are proficient at using speech recognition, you'll be able to produce text faster than most people can type. If you want, you'll also be able to control your computer handsfree because you'll be able to control your mouse and keyboard by voice.
Why is training needed to use Speech Recognition?
Speech recognition software is often sold by claiming that after a five-minute set up anyone can speak and see their words typed on the screen and control their computer by voice. While some people can use speech recognition without training, the reality is that most people need help to learn the skills necessary to produce satisfactory results consistently.
Speech recognition software works by using a speech file that is customized to your voice's pitch, tempo, volume and other characteristics. As you use speech recognition software, your speech file becomes either more accurate (smarter) or less accurate (dumber) depending on how you use it. If you don't specifically know how to make your speech file smarter, you can easily make it dumber. If your speech file gets dumber, your results get worse until you don't want to use it.
Without training, many people never understand why their words appear on the screen accurately sometimes and other times not, why their commands are executed sometimes and other times not.
After struggling to make speech recognition work reliably, the user either accepts that it only works partially for them or stops using it when they don't get the results they had hoped for. With training, these same people can learn to make their speech file smarter, get excellent results and be very happy that they have speech recognition.
Speech recognition software is an application of artificial intelligence that simulates the brain function responsible for speech recognition. The software manufacturers claim that it can be installed and used within minutes. Although it can be used within minutes after installation, the software has many nuances in its response to each specific client's speech pattern that noticeably affect accuracy, are not intuitive and often cause a client to believe that speech recognition technology is not as good as they were told it would be. With training, you won't be stopped by this.
Anyone who's ever used a computer knows that learning how to use a new software package is often not always as easy as advertised. Being trained exactly what to do is a lot better than being frustrated and struggling to figure out how to use what you've got.
Can I use Speech Recognition without training?
Some people are able to use speech recognition software without training, however, their numbers are few. Without training, many clients experience what appears to be unpredictable software behavior, have difficulty producing consistent results and quit in frustration. If you ever ask someone who unsuccessfully tried speech recognition software why they don't use it now, they will tell you "It didn't work for me".
Without training, most people don't understand why recognition is accurate some times but not others and few are persistent enough to continue using it. Understanding how the software responds under a range of conditions is critical for success.
There are different techniques for correcting recognition errors, editing text and navigating and controlling the mouse, keyboard and windows in one's computer. The most efficient technique to use depends on the particular circumstances at the moment. Before training, users do not know which technique is the most efficient to use for different circumstances, significant time is often wasted and the user is unaware of how much better they could do. Trained users will know which technique to use and when, making a large difference in user efficiency and satisfaction.
Individual training educates the user to clearly understand the software's response to their specific speech style under varying conditions, what commands to use when and how to use their software successfully without frustration to achieve their own and the software's full potential. With proper training, speech recognition software will work well for most people. Most people can learn the skills to be successful.
Why does Speech Recognition make errors?
There are a number of reasons why recognition errors occur. Factors that significantly affect results can be accents, speech volume, speed, annunciation and general speech style. Some words just sound like other words and speech recognition software may think you are saying other words. As some words have identical sounds but different spellings, recognition errors occur even under ideal conditions. Sometimes there are technical issues with the adjustments or settings of the microphone or sound card.
How does Speech Recognition work?
Speech recognition works by using artificial intelligence to recognize sounds, word usage frequency and word combinations. A speech recognition file records how you speak and becomes accustomed to your pronunciation and words that you use.
Imagine if you had a person learning how to speak English from you. This person gets smarter or dumber depending on the quality of your teaching. Speech recognition is similar. As you use speech recognition, your speech file either gets smarter or dumber.
Your training will teach you exactly how to make sure that your speech file always gets smarter and not dumber. With personal training, you can be sure that you will get the results you want -- Your results are guaranteed!
Sometimes, technical issues with hardware and/or software present insurmountable roadblocks for speech recognition clients. Requests for help at online speech recognition forums reveal that many people experience difficulty, confusion and frustration. This doesn't have to happen -- You don't have to be one of them! Included with your training is optimization of your hardware and software to make sure you get the best results.
Who is Speech Recognition for?
- People who want to produce writing without typing
- People who want to produce writing at greater than 100 words per minute
- Writers
- Professionals with limited time to write
- Students with lots of papers to write
- People who can't or don't want to use their hands to type or mouse
- People with writing challenges or learning disabilities
- People who want to control their computer by voice
What are the benefits of one-to-one individual training?
- You’re guaranteed to develop the skills to use speech recognition
- An experienced pro will customize your training for your specific needs
- You'll receive individualized guidance on how your specific speech pattern can be used to produce superior results
- Commands will be tuned to your voice giving you faster response time and higher efficiency
- You'll learn much more quickly than you could on your own – saving hours!
- You'll prevent frustration so you don’t quit
- You'll be treated like a human by another human – you'll receive the personalized attention you deserve and need to be successful
- Your configuration will be optimized to give you better noise cancellation to minimize environmental noise interference
What will I learn?
- How to speak to achieve maximum accuracy -- so the words you speak will be typed on your screen exactly as you say them as often as possible
- The best ways to correct recognition errors when they occur
- The fastest ways to edit your work when you want to rewrite/revise
- The techniques, tips and tricks to produce the results you want when you want them
Dan Multer did an excellent job training me on Dragon NaturallySpeaking. I learned each of the aspects in a focused, fun manner. I was never allowed to flounder and Dan Multer was always available, within one day to assist me again and again. I might have been able to set up Dragon on my own, but it was all of the extra information and knowledge which I received that made the difference for me. Knowledge that saved me countless hours of work and calls to the support desk. Those calls would have cost a bundle. I do not know what I would do without Dragon NaturallySpeaking. I could not have written this note without it. I have MS and am paralyzed. I am still able to work because of it. Thank you very much!
Rich Ursini - Branford, CT
I worked from my home office in Colorado as Dan trained me remotely from his office in Connecticut. Dan was an informed, helpful, and patient instructor. One of the most valuable things I experienced was his assistance in working through some of the program's technical challenges. These problems would have been very difficult for me to work through on my own, and may have forced me to give up. But, because of Dan's help, I was able to easily and effectively overcome them.
My typing speed is actually quite high - 80 words per minute. But by using this speech recognition program instead, I was able to dictate my stories even faster - at the rate of over 100 words per minute. As a writer, I believe it is the most valuable tool available today.
Lisa Kean – Louisville, CO
Optionally, you may decide that you also want to learn
- How to surf and search web sites, how to shop and bank by voice
- How to read and write e-mail and read and send attachments
- How to make entries into a calendar to schedule your time
- How to dictate into spreadsheets and databases
- How to control your mouse by voice and navigate around windows
- How to voice create, save, rename, and delete folders and files
- How to dictate when away from your computer using a mobile digital voice recorder and how to convert your recordings into text
What geographic areas do you cover?
I can train you online anywhere in the world -- we can connect computer to computer. We will be able to talk and hear each other and I will be able to see your screen the same as you do. If you prefer, I can visit you at your location in Connecticut, Rhode Island, Massachusetts, or New York. You may choose to visit me in Connecticut. I will work with you for what works best for your situation.
How long will it take to get trained?
Training time is determined by a number of factors such as your computer skills, speech style, and the speech recognition skills you want to acquire. A typical session lasts from one to two hours depending on your schedule and attention span.
Below are realistic, common training time ranges.
- Dictating -- speech style for accuracy, correcting, editing: 10-20 hours
- Internet Access -- surfing websites, searching, shopping, banking, social networks, accessing video, filling out forms: 2-4 hours
- E-mail -- reading, writing, attachments, contacts: 2-4 hours
- Computer Control -- Mouse, keyboard and Windows: 2-4 hours
- Mobile dictation using a Digital Voice Recorder: 2-4 hours
- Time management -- calendar, appointments, tasks: 2-4 hours
- Databases, spreadsheets -- entering data and formulas, navigating cells, fields: 2-4 hours
Does it matter whether I have Windows or Mac?
You can use speech recognition with either Windows or Mac. To use a Mac, you'll need additional software that allows Windows to run on the Mac. Just ask and I'll make sure you know what you need to have.
What languages do you train in?
I will train you in English. After you learn in English, you can use your skills in Spanish, German, French, Italian, or Dutch provided you are fluent in the additional language you would like to use.
How much will all this cost?
Dragon NaturallySpeaking offers the best results so that is the software I recommend. I offer training only, you may purchase the software from the vendor of your choice. If you want a vendor recommendation, I’ll be glad to make one.
Dragon is available from many online vendors. For many users, Dragon Naturally Speaking can be purchased for between $100 and $300 depending on the features and type of headset/microphone you want. If you want to dictate legal terminology, prices range from $500 to $800, the version to dictate medical terminology ranges from $1100 to $1600.
Training is $75/hour. Payments can be made for each session as you learn so you will only pay for what you need, want and receive. Depending on the skills you want to learn and your abilities, training time usually ranges between 10 and 20 hours and total cost will be between $750 and $1500.
The cost in time and money now is a small investment in yourself that will pay you back much more in future saved time, energy and independence. Targeted education to learn supporting skills for your specific needs can be invaluable.