LENA Designed for Live Speech As director of Child Language Research for the LENA Foundation, I commend the Louisiana State University (LSU) investigators involved with the LENA system research reported in the May 5 issue of The ASHA Leader. As the report highlights, the LENA system can make an important contribution to clinical practice. As ... Inbox
Free
Inbox  |   August 01, 2009
LENA Designed for Live Speech
Author Notes
Article Information
Development / Hearing & Speech Perception / Acoustics / Research Issues, Methods & Evidence-Based Practice / Professional Issues & Training / ASHA News & Member Stories / Normal Language Processing / Inbox
Inbox   |   August 01, 2009
LENA Designed for Live Speech
The ASHA Leader, August 2009, Vol. 14, 4. doi:10.1044/leader.IN1.14102009.4
The ASHA Leader, August 2009, Vol. 14, 4. doi:10.1044/leader.IN1.14102009.4
As director of Child Language Research for the LENA Foundation, I commend the Louisiana State University (LSU) investigators involved with the LENA system research reported in the May 5 issue of The ASHA Leader. As the report highlights, the LENA system can make an important contribution to clinical practice. As mentioned in the article, the investigators used pre-recorded archival data to evaluate the accuracy of the LENA system, and they report low reliability for conversational turns between the adult and child when compared with transcriptions.
We wish to point out that the LENA system was designed to record live interactions between the child and caregiver and was never intended for use with pre-recorded data. In fact, it is important to note that in order to correctly identify the child, our recording device must be within 6–8 inches of the child's mouth and it must record live speech. Due to the acoustic characteristics associated with recorded audio played from a speaker, the LENA system will generally mischaracterize that child's speech as speech belonging to someone else or as "electronic media." Thus, the relatively high error rates that the LSU researchers reported may be attributed to the fact that the child engaging in the turns could not be correctly identified by the LENA system because it was pre-recorded speech.
0 Comments
Submit a Comment
Submit A Comment
Name
Comment Title
Comment


This feature is available to Subscribers Only
Sign In or Create an Account ×
FROM THIS ISSUE
August 2009
Volume 14, Issue 10