 |
 |
 |
Confusion matrices for judges 2/3. Judge 2 is represented on the horizontal axis,
while Judge 3 is on the vertical. The axes represent the accent judgement, and the
values in the matrix are the number of file judgements at each vertex.
Arabic
112 utterances
1 2 3 4
1 6 7 6
2 2 16 27
3 8 35
4 3
BPortuguese
459 utterances
1 2 3 4
1 14 36 21
2 14 105 123 1
3 42 101
4 2
Cantonese
261 utterances
1 2 3 4
1 4 17 5
2 5 62 63
3 1 23 74 1
4 1
Czech
102 utterances
1 2 3 4
1 2 4 8
2 2 22 22
3 1 8 32
4 1
Farsi
261 utterances
1 2 3 4
1 3 7 3
2 3 43 81
3 24 89
4 3 1
French
284 utterances
1 2 3 4
1 9 27 12
2 8 52 79
3 1 29 61
4 1 2 1
German
326 utterances
1 2 3 4
1 16 28 16
2 8 78 73
3 1 23 70
4 2
Hindi
348 utterances
1 2 3 4
1 4 2 3
2 4 46 104
3 2 41 135
4 1 1
Hungarian
277 utterances
1 2 3 4
1 6 13 9 1
2 10 43 71
3 1 34 80 2
4 1 3
Japanese
194 utterances
1 2 3 4
1 7 6 6
2 3 38 49
3 1 17 63
4 2
Korean
169 utterances
1 2 3 4
1 4 8 4
2 2 24 45
3 22 57 1
4 1
Malay
56 utterances
1 2 3 4
1 1 1 2
2 8 17
3 5 21
4
Mandarin
282 utterances
1 2 3 4
1 4 19 6
2 54 72
3 1 30 92 1
4 1
Italian
214 utterances
1 2 3 4
1 8 12 5
2 4 36 45 1
3 1 18 68 3
4 1 4 1
Polish
143 utterances
1 2 3 4
1 6 11 4
2 1 16 29
3 15 56 1
4 2 2
Spanish
308 utterances
1 2 3 4
1 19 39 14
2 8 57 83
3 1 14 47
4 1 2
PPortuguese
66 utterances
1 2 3 4
1 2 5 2
2 2 11 15
3 8 21
4
Russian
236 utterances
1 2 3 4
1 2 19 7
2 4 53 65
3 1 12 67
4 4 1
Swedish
203 utterances
1 2 3 4
1 17 21 16
2 7 67 47
3 13 15
4
Swahili
71 utterances
1 2 3 4
1 1 1
2 1 17 17
3 1 10 22
4 1
Tamil
326 utterances
1 2 3 4
1 3 4 1
2 1 34 94
3 31 123 1
4 3 5 2
Vietnamese
134 utterances
1 2 3 4
1 2 2 7
2 2 18 30
3 12 43 1
4 3 1
ALL Languages
4961 utterances
1 2 3 4
1 153 301 163 1
2 92 923 1271 2
3 13 447 1416 11
4 10 42 8
|
 |
