CHECK/MLU error

32 views
Skip to first unread message

Sabrina Huang

unread,
Oct 2, 2018, 2:56:49 PM10/2/18
to chibolts
Hi,

I am trying to run a CHECK/MLU of a transcription of conversations in a preschool classroom.

When I try to run a check, this is the output:
> check FH2TDLL.cha
Language codes file being used: "Z:\CHILDES\CLAN\lib\fixes\ISO-639.cut"
Dep-file being used: "Z:\CHILDES\CLAN\lib\depfile.cut"
check FH2TDLL.cha
Tue Oct 02 14:53:57 2018
check (01-Aug-2018) is conducting analyses on:
  ALL speaker tiers
    and those speakers' ALL dependent tiers
  and ALL header tiers
****************************************

>

It doesn't say it is a success, but prior to this, the program was running correctly in telling me there were errors in my transcription.

Similarly, when I try to run a MLU, this is the output:


From file <FH2TDLL.cha>
MLU for Speaker: *OSC:
  MLU (xxx, yyy and www are EXCLUDED from the utterance and morpheme counts):
    Number of: utterances = 0, morphemes = 0

MLU for Speaker: *CAR:
  MLU (xxx, yyy and www are EXCLUDED from the utterance and morpheme counts):
    Number of: utterances = 0, morphemes = 0

MLU for Speaker: *FAT:
  MLU (xxx, yyy and www are EXCLUDED from the utterance and morpheme counts):
    Number of: utterances = 0, morphemes = 0

MLU for Speaker: *UNI:
  MLU (xxx, yyy and www are EXCLUDED from the utterance and morpheme counts):
    Number of: utterances = 0, morphemes = 0

MLU for Speaker: *AXE:
  MLU (xxx, yyy and www are EXCLUDED from the utterance and morpheme counts):
    Number of: utterances = 0, morphemes = 0

MLU for Speaker: *DAM:
  MLU (xxx, yyy and www are EXCLUDED from the utterance and morpheme counts):
    Number of: utterances = 0, morphemes = 0

MLU for Speaker: *SHE:
  MLU (xxx, yyy and www are EXCLUDED from the utterance and morpheme counts):
    Number of: utterances = 0, morphemes = 0



    TIER "%MOR", ASSOCIATED WITH A SELECTED SPEAKER,
         HASN'T BEEN FOUND IN THE INPUT DATA!
ADD -t%mor TO YOUR COMMAND TO ANALYZE JUST THE MAIN LINE



This is clearly incorrect because it says the number of utterances and morphemes are both 0 for all the speakers.
I was wondering if this was a problem because I am using an older version of CLAN?

Thanks so much for your assistance!

Leonid Spektor

unread,
Oct 2, 2018, 3:28:25 PM10/2/18
to chib...@googlegroups.com
Hi,

CHECK no longer report a success. You know that your transcripts do not have errors if CHECK does not report any errors.

MLU runs on %mor tier by default. Having %mor tiers produces more accurate results. I would suggest that you download appropriate MOR grammar using "File->Get MOR Grammar" menu and then run MOR command on your transcript files to have %mor tier added to them.

But, if you do not want to do that, then as the warning message at the end of the run of MLU command suggests just add "-t%mor" option to your MLU command line and then MLU will run on speaker tiers instead.


Leonid.

--
You received this message because you are subscribed to the Google Groups "chibolts" group.
To unsubscribe from this group and stop receiving emails from it, send an email to chibolts+u...@googlegroups.com.
To post to this group, send email to chib...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/chibolts/924dacf3-1230-4f40-aede-0eccf722774f%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply all
Reply to author
Forward
0 new messages