My mother was recently diagnosed with ALS. She is still able to speak, but you can see from slight changes in her walking that the ALS is beginning to affect her. Even her speech, though fully understandable is slightly different. We would like to record her voice for a time when she may no longer be able to be understood. Is voice banking the best way to do that?
— Annie, Middlesex, NJ
Before discussing what’s best, let’s clarify some terminology. Augmentative Communication devices typically use a computer generated voice. Male, Female, even children’s voices are available. The quality of these voices have improved immensely over the years. Voice banking is the process of storing one’s own voice. So when an augmentative communication device is used, the voice you hear is the voice of the person who previously recorded their voice.
There are two voice banking options. One is recording your voice outright. ‘How are you’, ‘Let’s go to the park’, ‘lift my leg, please’. You would anticipate the messages, names, phrases, you want stored, and record each one. This is considered Digitized Speech. The 2nd Voice banking option is Synthesized Speech. You create a computer file of all the sounds your voice makes from the alphabet and combinations of letters. These sounds then are ‘synthesized’ (by a computer program) to form words, using each sound from your voice.
For the outright recording, digitized speech, words and phrases are recorded on a computer, and saved as a .wav or .mp3 file. You then can program each file to a specific location on the screen of a computer, and you have created a touch screen with multiple .wav or .mp3 recordings. One screen can store and play multiple recordings. This is an easier way to go. Simply record the words and phrases you want over time. The limitation is that it is not a natural way of communicating. Our language is complex and we constantly interject unique words and phrases.
Sound Recorder on Windows is a simple way to store your own recordings.
Audacity is a more complex program for recording words and phrases.
The Synthesized method takes more time. ModelTalker is the primary program used to create one’s own voice file. It request 4-6 hours of speaking specified sentences into the computer. The program then creates a voice file that can be used on some augmentative communication devices. The versatility of this method is that anything you spell on the device, pre-programmed or unique is now spoken with the voice sounds of the person.
What’s best? It’s really up to you. Know that the communication devices as they come now have very understandable and good quality voices. If it is important to have a device use your own voice, look into ModelTalker. That will give you the most versatility. If you are not up for several hours of recording, use the digitized recording methods and identify specific phrases you want to record.
Please contact me with any questions about voice banking or augmentative communication.
Ben Lieman, ATP, MSW is the Assistive Technology Specialist with the Greater New York Chapter, advising patients and caregivers about medical equipment, home accessibility, and augmentative communications devices. To ask Ben a question, simply email him at firstname.lastname@example.org or call at (212) 720-3057. Ben will answer all questions directly as usual, but not all questions will appear in the Monthly Update.