LENA

[Brian MacWhinney]

Dear Kim,
    This is very helpful.  This level of functionality for LENA is  
similar to
that reached in the  BlitzScribe program developed at the MIT
Media Lab. Unlike BlitzScribe, however, it appears that LENA is
ready for deployment by end users outside of a complex operating
system environment.  And the output of .cha files is a big plus.

   If you take a look at the goals for the CHILDES project renewal,
you will see that the inclusion in CLAN of methods for locating  
segments to
be transcribed is a major goal.  Whether it be through the LENA tools,  
the MIT
tools, or something new but similar, I certainly agree that there is a
pressing need to extend
current CLAN transcription to include a first-pass pre-linkage that  
can then
serve as the basis for a first-pass transcription.

-- Brian MacWhinney

On Feb 11, 2008, at 9:59 PM, Kim Oller wrote:

>
> Dear Nan and others interested in LENA,
> I'm on the scientific advisory board for Infoture, which produces
> LENA. The system locates utterances of child and adult, but does not
> identify words or phonetic elements - I am using LENA to help with
> transcription because it helps me find utterances fast - but the
> transcription itself still has to be done by a human. The big
> advantage of LENA is that the parent can conduct recording
> naturalistically in the home with substantial convenience for as much
> as 16 hours, (the recorder is battery powered) and then you have a
> huge sample of child and adult utterances to work with, all recorded
> outside the laboratory. I am currently working on just 8 days of
> recording with my daughter Lia, most of which were in the ten-hour
> range. Recording quality is quite good, as long as you don't have the
> TV or stereo on too close to the child - the software does a pretty
> good job of recognizing the TV and stereo, though, so the processed
> data is still quite useful.
>
> We are hoping that in the not too distant future, the LENA system will
> also yield a first level (approximate) transcription of child
> utterances, and first level identification of words for both child and
> adult.
>
> Just one more point on the LENA software; we are beta testing it for
> Infoture right now, and a key feature is the export of LENA labeled
> files (indicating both child and adult utterances) to .cha files for
> use with CLAN software. I'll be able to give a reading on how it works
> soon, and if there are improvements needed, we'll let them know.
> Hopefully it will be in good shape not long from now, and I expect it
> to be a great help in transcription.
> Best wishes,
> Kim
>
>
> On Feb 8, 9:04 am, "Nan Ratner" <nrat... at hesp.umd.edu> wrote:
>> At a conference a short while ago, I received a brochure regarding  
>> a program for child language recording and transcription called  
>> LENA. It supposedly automates a large portion of transcription for  
>> child language; has anyone on the list seen this program in  
>> operation, and how well does it appear to work?
>>
>> best regards to all,
>>
>> Nan Ratner
>>
>> Nan Bernstein Ratner, Professor and Chairman
>> Department of Hearing and Speech Sciences
>> 0100 Lefrak Hall
>> University of Maryland
>> College Park, MD 20742
>> nrat... at hesp.umd.eduhttp://www.bsos.umd.edu/hesp/facultyStaff/ratnern.htm
>> 301-405-4213
>> 301-314-2023 (fax)
> >
>


--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Info-CHILDES" group.
To post to this group, send email to info-childes at googlegroups.com
To unsubscribe from this group, send email to info-childes-unsubscribe at googlegroups.com
For more options, visit this group at http://groups.google.com/group/info-childes?hl=en
-~----------~----~----~----~------~----~------~--~---