md_test4.ref.rttm
3.68 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
;; mdeval md_test4 ref rttm
;; This test file has some odd speaker names,
;; and will test that md-eval counts Nref for
;; "SU exact detection statistics in terms of words"
;; the same as Nref for "SU detection in terms of # of SU's"
;; This is an important test of getting that logic correct.
SPKR-INFO md_test4 1 <NA> <NA> <NA> child spkr1r <NA>
SEGMENT md_test4 1 10.0 6.0 <NA> <NA> spkr1r <NA>
SPEAKER md_test4 1 10.0 6.0 <NA> <NA> spkr1r <NA>
SU md_test4 1 10.0 3.0 <NA> statement spkr1r <NA>
LEXEME md_test4 1 10.0 1.0 firstWord lex spkr1r <NA>
LEXEME md_test4 1 11.0 1.0 T.'s alpha spkr1r <NA>
LEXEME md_test4 1 12.0 1.0 thirdWord acronym spkr1r <NA>
SU md_test4 1 13.0 3.0 <NA> question spkr1r <NA>
LEXEME md_test4 1 13.0 1.0 fourthWord interjection spkr1r <NA>
LEXEME md_test4 1 14.0 1.0 fifthWord propernoun spkr1r <NA>
LEXEME md_test4 1 15.0 1.0 sixthWord other spkr1r <NA>
;; Can the scanner handle this oddly formatted speakername?
SPKR-INFO md_test4 1 <NA> <NA> <NA> unknown name="<NAr>" <NA>
SEGMENT md_test4 1 16.0 3.0 <NA> <NA> name="<NAr>" <NA>
SPEAKER md_test4 1 16.0 3.0 <NA> <NA> name="<NAr>" <NA>
;; In the sys file, the last word of this SU will be omitted
SU md_test4 1 16.0 3.0 <NA> backchannel name="<NAr>" <NA>
LEXEME md_test4 1 16.0 1.0 seventhWord lex name="<NAr>" <NA>
LEXEME md_test4 1 17.0 1.0 eighthWord lex name="<NAr>" <NA>
LEXEME md_test4 1 18.0 1.0 ninthWord lex name="<NAr>" <NA>
;; How about this speakername?
SPKR-INFO md_test4 1 <NA> <NA> <NA> unknown spkr=<sp3r> <NA>
SEGMENT md_test4 1 19.0 11.0 <NA> <NA> spkr=<sp3r> <NA>
SPEAKER md_test4 1 19.0 11.0 <NA> <NA> spkr=<sp3r> <NA>
;; In the UEM file, the last 1.51 words of this SU will be no-eval'd
;; and in the sys file the last two words of it will be omitted
SU md_test4 1 19.0 5.00 <NA> question spkr=<sp3r> <NA>
LEXEME md_test4 1 19.0 1.0 tenthWord lex spkr=<sp3r> <NA>
IP md_test4 1 20.0 <NA> <NA> filler spkr=<sp3r> <NA>
FILLER md_test4 1 20.0 1.0 <NA> filled_pause spkr=<sp3r> <NA>
LEXEME md_test4 1 20.0 1.0 eleventhWord fp spkr=<sp3r> <NA>
LEXEME md_test4 1 21.0 1.0 twelfthWord lex spkr=<sp3r> <NA>
LEXEME md_test4 1 22.0 1.0 thirteenthWord lex spkr=<sp3r> <NA>
LEXEME md_test4 1 23.0 1.0 fourteenthWord lex spkr=<sp3r> <NA>
;; The following ref SU is a duplicate of the preceding ref SU
;; In the UEM file, the last 1.50 words of this SU will be no-eval'd
;; rather than the last 1.51
;; and in the sys file the last two words of it will be omitted
SU md_test4 1 24.0 5.00 <NA> question spkr=<sp3r> <NA>
LEXEME md_test4 1 24.0 1.0 fifteenthWord lex spkr=<sp3r> <NA>
IP md_test4 1 25.0 <NA> <NA> filler spkr=<sp3r> <NA>
FILLER md_test4 1 25.0 1.0 <NA> filled_pause spkr=<sp3r> <NA>
LEXEME md_test4 1 25.0 1.0 sixteenthWord fp spkr=<sp3r> <NA>
LEXEME md_test4 1 26.0 1.0 seventeenthWord lex spkr=<sp3r> <NA>
LEXEME md_test4 1 27.0 1.0 eighteenthWord lex spkr=<sp3r> <NA>
LEXEME md_test4 1 28.0 1.0 nineteenthWord lex spkr=<sp3r> <NA>
SU md_test4 1 29.0 1.0 <NA> question spkr=<sp3r> <NA>
LEXEME md_test4 1 29.0 1.0 twentiethWord lex spkr=<sp3r> <NA>