The dataset consists of words in thee quantity degrees with CVC(C)V, VCCV structure, read by 6 male and 6 female native Estonian subjects. The test words were in the middle and end of 4-word carrier sentences where all the words had 2 syllables. 
The data set was collected for the bachelors thesis: "The quality and quantity of the intervocalic /l/" (Anton Malmi, 2014)

The data set has the following info:

KJ - The sex and the number of the test subject. Male (m) or female (f)
Fail - The number of the file that the data was extracted from. It ranged from 1 to 12, denoting the number of sentences that the subject had to read
Sugu - The sex of the test subject. Male (m) or female (f)
Pos - The position of the testword in the utterance. Utterance-medial (LK) and final (LL)
Valde - Quantity - Quantity degree as a number, levels 1/2/3
Vokaal - The vocalic context in which the consonant /l/ was environed. Levels /a e i u/
V1kest - The duration of the vowel preceding /l/ in milliseconds
C2kest - The duration of the intervocalic /l/ in milliseconds
V2kest - The duration of the vowel after /l/ in milliseconds
V1F1_10 to V1F1_100 - The first formant values (in Hz) of the vowel preceding /l/ with a 10% step from the whole duration
V1F2_10 to V1F2_100 - The second formant values (in Hz) of the vowel preceding /l/ with a 10% step from the whole duration
C2F1_10 to C2F1_100 - The first formant values (in Hz) of /l/ with a 10% step from the whole duration
C2F2_10 to C2F2_100	- The second formant values (in Hz) of /l/ with a 10% step from the whole duration
V2F1_10 to V2F1_100 - The first formant values (in Hz) of the vowel after /l/ with a 10% step from the whole duration
V2F2_10 to V2F2_100 - The second formant values (in Hz) of the vowel after /l/ with a 10% step from the whole duration