Link to Text Version Speech Research Lab
design bar
Speech Research Info
Link to Speech Research Lab Main Page
Link to info for users of Augmentative and Alternative Communication Devices
Link to info for Clinicians
Link to info for Speech Researchers
Link to Model Talker Speech Synthesizer
Link to STAR : Speech Training, Assessment and Remediation
Link to Language and Speech Journal
Link to Speech Research Lab Newsletters
Contact Us!

 

 

 

 

 

Database of Children's Speech

The Speech Research Laboratory at the Alfred I. duPont Hospital for Children is undertaking a project to gather voice recordings of 500 young children between the ages of six and eight, whose native language is American English. The recording process, including positioning the microphone, providing instructions, and recording a 100-item word list is expected to last approximately fifteen minutes for each child. The Speech Research Laboratory plans to use these recordings in a number of ways.

One application is with our Speech, Training, Assessment, and Remediation (STAR) project.  Over the past few years we have been developing an interactive computer game called STAR.  This game provides children with a structured opportunity to practice producing a variety of speech sounds and gives them feedback on whether they are producing the sounds correctly. In order to continue to make improvements in the speech recognition technology that this program uses, we are seeking a collection of voice recordings for analysis of typically developing childrens’ speech patterns.

Another application is with our ModelTalker project.  The Speech Laboratory has developed a software program that provides a computer generated synthetic voice for individuals who have difficulty speaking.  Adult subjects are currently testing this program.  Collecting and analyzing typically developing children's speech will help our researchers consider ways to modify this program so that children who are unable to speak might one day use it to generate a naturally sounding age-appropriate synthetic voice.

The development of this speech database is supported by Voiceware Co., Ltd. In addition to our use of the speech data for research, Voiceware will use the data in their product development laboratories. The Speech Research Laboratory will not be involved in their product development projects other than providing a copy of the collection of voice recordings that we are gathering for our database. Children's identifying information (name, age, and place of residence) will not be provided to Voiceware. There will be no charge or financial compensation for participation in the voice recording process.

Projects | Publications | Related Links | Staff |
Facilities | Events | Positions