SpeakQ (English)

What can I do with SpeakQ?

You can do everything that WordQ does, plus you can write by speaking. At any time you have the choice of typing with the keyboard, using word prediction, or speaking straight into your text. Speech recognition and word prediction are integrated to enhance the effectiveness of each other.

You can train SpeakQ to recognize your speech using texts provided that match your reading level or you can write your own training texts. An innovative training interface makes training simple. Just listen to the computer speak and repeat after it!:

Once you have finished training, you can begin dictation using one of two dictation modes:

In the Speak and select mode spoken words are displayed as a list of choices enhanced by word predictions. You then select from the list of words or phrases (using the keyboard or mouse), or type letters to further refine the suggestions.

In the Speak continuously mode your spoken words are typed directly into your document as you say them.

Recognized words and phrases are immediately read aloud to help you identify mis-recognitions. To help you make any corrections, highly accurate word predictions are suggested by combining speech with predictive information. As a result, the desired word can be selected with a minimum number of keystrokes.

It may take you some time to work out for yourself how to balance typing with speaking. Everyone is different.

Key Benefits

• Enjoy all of the existing benefits of WordQ
• Train speech recognition using training text of your choice
• Train using speech prompting if you have difficulty reading the training text
• Work directly within any standard application (e.g. Microsoft Word, Wordpad, Outlook)
• Dictate either continuously into a document or discretely into the word prediction box
• Dictate in phrases or individual words
• Hear what the computer recognizes
• Type, use word prediction, or speak-to-write at any point in your writing
• Make corrections using speech-enabled word prediction
• Back-up and restore your speech profile
• Use speech recognition without having to learn and memorize speech commands

SpeakQ v1.2 Improvements

• Microsoft English (U.S.) v6.1 Recognizer is now included with SpeakQ. Previously, this advanced recognition engine was only available if Office 2003 was installed with the Speech Recognition option.
• An option has been added to permit users in school settings to store their WordQ vocabulary files and SpeakQ speech speech files on a central server instead of the local workstation. This capability permits a student to move from workstation to workstation and have access to their WordQ/SpeakQ settings available wherever they log in. Network administrators can find out more about enabling this option by viewing the FAQ WordQ v2.2 and SpeakQ v1.2 Network Installation
• The Acapela text-to-speech engine has been upgraded to address a potential incompatibility with other applications using Acapela's text-to-speech products.

Background

SpeakQ was developed by a research and development team at Bloorview MacMillan Children's Centre in collaboration with the Education Development Center (Newton, MA) with funding from the National Institute of Disability and Rehabilitation Research, US Department of Education Grant No. H133G000204-02 as well as the Ontario Rehabilitation Technology Consortium.

Speech recognition technology enhances the writing abilities of many individuals, yet some individuals who may benefit from this technology, do not pursue its use. In the case of some students with learning disabilities, present-day speech recognition products do not meet their needs due to reading demands, operational complexity, and the lack of integration with assistive technologies such as word prediction and text-to-speech. This has led to the development of SpeakQ, an innovative new software program that combines the power of speech recognition with concurrent access to other technologies, and offers students a writing solution with distinct advantages.

SpeakQ enables users to dictate words into any standard Microsoft Windows document in either a continuous or discrete word-by-word manner. SpeakQ is used in combination with WordQ 2, a writing tool with advanced word prediction and text-to-speech capabilities. Together, these tools provide an integrated and flexible approach to writing, and assist students with their spelling, proofreading, and editing activities. Students now have the option to dictate text, type on their keyboard, choose word predictions from a list, or alternate between these methods, each of which has been optimized to enhance the effectiveness of the other.

Designed with ease-of-use in mind, SpeakQ simplifies speech recognition technology. During the training (or enrollment) process, SpeakQ automatically reads the required training text to users, who then need only to repeat the spoken words or phrases at their own pace. At the same time, the training display minimizes visual distractions, and accommodates users with visual differences. To reduce the complexity of dictation, SpeakQ emphasizes the task of writing, and eliminates the need to memorize voice commands. In addition, recognized words and phrases are immediately read aloud to facilitate the detection of recognition errors. Then, to facilitate error correction, highly accurate word alternatives are suggested by combining speech data with statistical predictive data. As a result, the desired word can be selected with a minimum number of keystrokes.

SpeakQ supports two styles of dictation. Like other speech recognition products, it allows users to dictate text directly into their documents. In addition, it also supports a style of dictation that is unique to SpeakQ. This approach uses word prediction capabilities and introduces a concept we call “speech-enabled word prediction.” In effect, spoken words are displayed as a list of choices enhanced by word predictions. Users select from the list of words, or type letters to further refine their choices. By using this approach, not only can students control the words that are entered into their documents as they speak, but they can also improve the accuracy of the words. Consequently, they can spend less time correcting recognition errors, and more time on the real task of writing.