Error-free verbs

[May Y. Wu]

Dear all,

We are trying to calculate the ratio of error-free verbs in our transcripts. The transcripts look like this:

*XXX: &-and &-uh after that she continue [* st] [* m:a]
[^c id] to (.) &-um (.) be <a actress> [//] an actress [^c nf]
and (.) give [* st] [* m:a] us (.) several (.) very (.) significant
(.) roles [^c d] .
%mor: prep|after comp|that pro:sub|she v|continue inf|to cop|be det:art|a
n|actress coord|and v|give pro:obj|us qn|several adv|very
adj|significant n|role-PL .

We understand from the CLAN manual that it is possible to use "freq" to search for " v|*" in the %mor tier, and then switch back to the speaker's tier to see if the corresponding verb is followed by any errors "[\* *]". However, we did not succeed. We have tried 

freq alt1.cha +sm|v +d7

and

kwal alt2.cha +t%mor +sv* +d31

Do you have any suggestions on how we should do it?

Thanks ahead for any insights!

--

May

[Leonid Spektor]

May,

You can not do this in one command. Try the following commands and see if they help:

To get all verbs: freq +sm|v filename.cha

To get all error verbs: freq +sm|v,** filename.cha

To get mix of both: freq +sm|v +sm** filename.cha

These commands are just for more example:

freq +sm|v,o% test.cha +d6

freq +sm|v,**,o% test.cha +d6

freq +sm|v,*%,o% test.cha +d6

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 view this discussion on the web visit https://groups.google.com/d/msgid/chibolts/43903434-eba0-4646-b35e-39f8e3dc1abdn%40googlegroups.com.