Voice Recognition (Speech to Text Software)

Description

an image of a earphone and microphone headset

 

 

 

 

Voice recognition software (also known as speech to text software) allows an individual to use their voice instead of typing on a keyboard. Voice recognition may be used to dictate text into the computer or to give commands to the computer (such as opening application programs, pulling down menus, or saving work). Voice recognition software allows for a quick method of writing. It is also useful for people with disabilities who find it difficult to use the keyboard. This software can also assist those who have difficulty with transferring ideas onto paper, as it helps take the focus out of the mechanics of writing (e.g., spelling, sentence structure, etc.). Not all speech recognition software packages are equal in function, capability, or ease of use.

Early voice recognition applications required each word to be separated by a distinct space (i.e., briefly pausing between each word). This allowed the software to determine where one word begins and the next stops. This style of dictation is called discrete speech. Few people prefer these systems to the newer continuous speech model. It is hard to find discrete speech recognition software due to advancements in continuous speech recognition software. Continuous speech voice recognition applications allow a user to dictate text fluently into the computer.

All speech recognition software requires a sound card and microphone; manufacturers recommend specific brands and models that have been tested. The microphone captures the user's speech and the sound card converts the speech into a digital form that the software can interpret. This digital representation is compared to a profile of the user's speech. The software creates this profile by having the user initially go through a training exercise in which they are asked to speak specific sentences and words. By comparing spoken words with the user profile, the software determines which words the user probably spoke. The software also uses a general user profile (based on speech from many people) and grammar rules and relationships between words to improve recognition. While the accuracy of voice recognition has improved significantly over the past few years, some users still experience problems with accuracy either because of the way they speak or the nature of their voice.

Digital voice recorders are frequently used with voice recognition software. These allow the user to dictate text into the voice recorder which can be translated at a later date by a voice recognition system.

Voice recognition software is becoming increasingly popular as a built-in function, especially in tablets and smartphones. Comparing the recognition and ease of use may help determine if built-in is good enough or if the student requires a more robust solution, especially for writing longer blocks of text.

Want to learn more about speech to text software? Ottawa Network for Education created videos for students who are interested in learning about or starting to use technologies for reading, writing and other school work. To learn more about speech to text software, visit Ottawa Network for Education's website to watch a video titled ‘AT Tools for Writing’ and 'Tools for Writing and Editing.'

 

Points to Ponder: Questions to consider when choosing Voice Recognition Software

  • Features
    • Can the software be used for both dictation and control of the computer or device?
    • What languages and accents does it support?
    • How fast is it at translating your voice into words or actions?
    • Does it have macro compatibility? (i.e., can you program simple commands that can be used to type long blocks of text or execute more complicated series of operations)? Sometimes macros can be used to control other types of technology (e.g., magnification keyboard shortcuts)
    • Does it allow for wireless dictation, i.e., Bluetooth compatible?
    • Does it allow you to import/export word lists and user profiles?
    • Is portability required? Does it allow for transcription from a digital voice recorder? Consider choosing a model compatible with the voice recognition software.
    • Does it need to work with other technology (e.g., screen reading technology for users with low vision) and are there compatible options or bridge programs (i.e. programs that make the two work together)?
    • Usability?
      • How comfortable is the user typing using their voice? Can they compose their thoughts or commands while the microphone is on?
      • What is the learning curve of the software? (i.e. how quickly will it learn your voice, do you need to learn how to “correct” or “train” the program to understand your voice?) How easy is it to program new words or names?
      • How forgiving is the software for variations in speech (e.g., unique accents, stuttering, limited breath, etc.)?
      • How does it provide feedback that the microphone is on or that the words have been typed correctly? (e.g., visual, auditory)
      • Is there an option to have dictated text read aloud?
      • Does it have a user-friendly interface (i.e. menu, toolbar, control panel)? How complex is the menu or toolbar structure? How easy is it to adjust settings? How easy is it to program custom commands if available?
      • Compatibility
        • Is it compatible with your computer or device's operating system? Does it integrate with existing applications (e.g., your word processing program, collaborative platforms, internet browsers)?
        • Does it ship with its own microphone?
        • Can the student position their own microphone?
        • Will you need a USB microphone (since it comes with its own sound card)?
        • Are there additional features or custom options that match your users’ needs? (e.g., adjusting the visual feedback, choices, or lists by displayed size, font, or adding auditory feedback)
        • Support
          • What type of built-in tutorial are available? What other supports are available for users?
          • Is there a trial period or demo copy for testing out the software?
          • How will you get support if you need it? (e.g., a technician in the school, a local vendor, by telephone, by email, remote access, etc.)
          • How much do upgrades cost? Is there any software management agreement or upgrade package that can be purchased?
          • What is the payment model? (e.g., purchase, pay per use)

 

Resources

Manufacturers of Voice Recognition Software

SNOW does not endorse any of the following software/hardware. These links are provided for information purposes only.

Nuance  – Dragon Naturally Speaking - Home, Premium, and Individual versions (Windows); Dragon Individual (Mac); also has various medical, legal, and other specialized options

goQSoftware. – SpeakQ (requires WordQ) (Windows)

Software Programs that use Voice Recognition Technology

iCommunicator – older software that uses voice recognition technology to convert spoken language into sign language for those who are deaf and hard of hearing.

Metroplex Voice Computing, Inc. – has developed hands/mouse free software which utilizes voice recognition technology to dictate math.

Trigram Technology – has developed software to be used with voice recognition technology including Language Models that assist those using medical terminology during dictation. This software helps to increase accuracy rates

 

Free/Open Source Software

Gnome Voice Control (Linux)

Open Mind Speech (Linux)

Perlbox (Linux)

 

Built-In Voice Recognition

Did you know? That your computer or device may have free built-in speech recognition software available. Learn more about the INSERTLINKLATER accessibility features of your computer or device here.

 

Additional Useful Links

Say I Can – provides tips and instructional videos for using voice recognition software