Voice Banking is a process of recording a large inventory of your speech that is then used to create a synthetic voice that approximates your natural voice.
Done successfully, this would allow one to spell and create unique messages and then speak them through a synthesizer that approximates one’s natural speech. The science behind this process continues to
be in development with beta-versions of available software. The ModelTalker is
one such project from the University of Delaware Speech Research Lab. The website is:
www.asel.udel.edu/speech/ModelTalker.html
Message Banking with your own voice
digitally record and store words, phrases, sentences, personally meaningful sounds and/or stories using your natural voice, inflection and intonation.
These messages are catalogued as .wav files and may then be linked to messages in a variety of augmentative communication technologies or sound storage files. This will allow you to ‘retrieve’ a message and
speak it in your own voice but does not allow you to create novel messages by spelling. If you have recorded individual words, you may combine those words to create unique messages, although the output will sound more staccato than your natural speaking.
Message Banking by proxy
is the election of a proxy voice to do all recordings because issues of fatigue, pain or intelligibility may make it difficult for an individual to bank all messages. In this case, the patient may still bank ‘legacy
messages’ but the majority will be banked by proxy.
781 216 2252 fax
www.childrenshospital.org/acp
FACEBOOK: www.facebook.org/acpchboston