Email Print Share
May 2, 2017

Custom, digital voices for people unable to speak -- VocaliD

More than 10 million people worldwide are unable to speak and more than 500 million devices and applications are voice-narrated. All of them choose from the same few generic, mechanical voices, according to VocaliD, a small business funded by the National Science Foundation (NSF).

VocaliD creates custom digital voices by taking a tiny sample of someone’s voice--like sustained "awe"--and blending it with a speaker who is about the same age and size with the same linguistic background. Rupal Patel, CEO and founder of VocaliD, says that they merge the samples of speech together and create a unique voice that sounds more like the person's voice.

VocaliD is supported by NSF's Small Business Innovation Research (SBIR)/Small Business Technology Transfer (STTR) program, a nearly $190 million program that awards research and development grants to small businesses and startups, transforming scientific discovery into products and services with commercial and societal impact. To learn more visit:

Credit: NSF

Images and other media in the National Science Foundation Multimedia Gallery are available for use in print and electronic material by NSF employees, members of the media, university staff, teachers and the general public. All media in the gallery are intended for personal, educational and nonprofit/non-commercial use only.

Videos credited to the National Science Foundation, an agency of the U.S. Government, may be distributed freely. However, some materials within the videos may be copyrighted. If you would like to use portions of NSF-produced programs in another product, please contact the Video Team in the Office of Legislative and Public Affairs at the National Science Foundation.

Additional information about general usage can be found in Conditions.