Research Behind Everything We Do
How We Began
Inspired by research on child development, especially Meaningful Differences by Drs. Hart and Risley, the LENA Research Foundation's world-class team of engineers, research academicians, linguists, and speech-language professionals spent years researching and developing the LENA System.
A simple technology to use, the LENA System was complex for our scientists to develop. The LENA Digital Language Processor (DLP) captures every utterance between adults and children, as well as all other language environment audio. Then the software uses advanced algorithms and statistical models of adult speech and child vocalizations to analyze the language environment.
LENA Software Development
To create the LENA software, LENA Research Foundation scientists constructed algorithms and statistical models for adult speech and child vocalizations utilizing over 18,000 hours of recordings from 314 families. The recordings were repeatedly fed into a specially designed super computer with 172 parallel processors and 27,300 gigabytes of storage.
In a nutshell, LENA analyzes every millisecond of the audio file and produces core reports as well as segmented data for researchers to conduct additional analysis. The core reports provide counts for Adult Words, Conversational Turns, Child Vocalizations, as well as normative comparison information, automatic vocalization assessments, developmental age, and a breakdown of the components in the audio environment.
LENA filters out far field speech and estimates the number of words spoken by any adult — mother, father, or other caregiver — near the child.
A conversational turn is considered to have occurred when a child vocalizes and an adult responds, or an adult speaks and a child responds. Each time that happens, one turn is counted.
One child vocalization is counted when a child’s speech sound of any length is surrounded by at least 300 milliseconds of silence.
LENA also identifies and segments the majority of non-speech sounds a child makes. Such sounds include vegetative sounds — those that have to do with respiration or digestion — as well as cries, screams, and whining sounds.
The LENA Language Environment software is particularly unique in that our engineers adapted cutting-edge speech recognition software to work in unstructured environments. Doing so enables LENA to eliminate child crying and vegetative sounds, as well as voices from a radio or television.