sastt-case1.sys.rttm.filt.mdeval 6.97 KB
command line (run on 2009 May 11 at 13:31:08) Version: 22  ../../md-eval/md-eval.pl -nafcs -c 0.25 -o -r sastt-case1.ref.rttm.filt -s sastt-case1.sys.rttm.filt -M sastt-case1.sys.rttm.filt.mdeval.spkrmap

Time-based metadata alignment

Metadata evaluation parameters:
    time-optimized metadata mapping
        max gap between matching metadata events = 1 sec
        max extent to match for SU's = 0.5 sec

Speaker Diarization evaluation parameters:
    The max time to extend no-score zones for NON-LEX exclusions is 0.5 sec
    The no-score collar at SPEAKER boundaries is 0.25 sec

Exclusion zones for evaluation and scoring are:
                             -----MetaData-----        -----SpkrData-----
     exclusion set name:     DEFAULT    DEFAULT        DEFAULT    DEFAULT
     token type/subtype      no-eval   no-score        no-eval   no-score
             (UEM)              X                         X
         LEXEME/un-lex                    X                          
        NON-LEX/breath                                              X
        NON-LEX/cough                                               X
        NON-LEX/laugh                                               X
        NON-LEX/lipsmack                                            X
        NON-LEX/other                                               X
        NON-LEX/sneeze                                              X
        NOSCORE/<na>            X         X               X         X
 NO_RT_METADATA/<na>            X                                    
             SU/unannotated               X                          

*** Performance analysis for Speaker Diarization for c=1 f=ICSI_20011030-1030_d*_NONE ***

    EVAL TIME =     30.00 secs
  EVAL SPEECH =     30.00 secs (100.0 percent of evaluated time)
  SCORED TIME =     28.50 secs ( 95.0 percent of evaluated time)
SCORED SPEECH =     28.50 secs (100.0 percent of scored time)
   EVAL WORDS =     14        
 SCORED WORDS =     14         (100.0 percent of evaluated words)
---------------------------------------------
MISSED SPEECH =      0.00 secs (  0.0 percent of scored time)
FALARM SPEECH =      0.00 secs (  0.0 percent of scored time)
 MISSED WORDS =      0         (  0.0 percent of scored words)
---------------------------------------------
SCORED SPEAKER TIME =     28.50 secs (100.0 percent of scored speech)
MISSED SPEAKER TIME =      0.00 secs (  0.0 percent of scored speaker time)
FALARM SPEAKER TIME =      0.00 secs (  0.0 percent of scored speaker time)
 SPEAKER ERROR TIME =      0.00 secs (  0.0 percent of scored speaker time)
SPEAKER ERROR WORDS =      0         (  0.0 percent of scored speaker words)
---------------------------------------------
 OVERALL SPEAKER DIARIZATION ERROR = 0.00 percent of scored speaker time  `(c=1 f=ICSI_20011030-1030_d*_NONE)
---------------------------------------------
 Speaker type confusion matrix -- speaker weighted
  REF\SYS (count)      adult_male            MISS              
adult_male                1 / 100.0%          0 /   0.0%
  FALSE ALARM             0 /   0.0%
---------------------------------------------
 Speaker type confusion matrix -- time weighted
  REF\SYS (seconds)    adult_male            MISS              
adult_male            28.50 / 100.0%       0.00 /   0.0%
  FALSE ALARM          0.00 /   0.0%
---------------------------------------------

*** Performance analysis for Speaker Diarization for c=1 f=VT_20051027-1400 ***

    EVAL TIME =      7.50 secs
  EVAL SPEECH =      7.50 secs (100.0 percent of evaluated time)
  SCORED TIME =      5.40 secs ( 72.0 percent of evaluated time)
SCORED SPEECH =      5.40 secs (100.0 percent of scored time)
   EVAL WORDS =      9        
 SCORED WORDS =      7         ( 77.8 percent of evaluated words)
---------------------------------------------
MISSED SPEECH =      0.75 secs ( 13.9 percent of scored time)
FALARM SPEECH =      0.00 secs (  0.0 percent of scored time)
 MISSED WORDS =      0         (  0.0 percent of scored words)
---------------------------------------------
SCORED SPEAKER TIME =      6.80 secs (125.9 percent of scored speech)
MISSED SPEAKER TIME =      2.15 secs ( 31.6 percent of scored speaker time)
FALARM SPEAKER TIME =      0.25 secs (  3.7 percent of scored speaker time)
 SPEAKER ERROR TIME =      0.00 secs (  0.0 percent of scored speaker time)
SPEAKER ERROR WORDS =      3         ( 42.9 percent of scored speaker words)
---------------------------------------------
 OVERALL SPEAKER DIARIZATION ERROR = 35.29 percent of scored speaker time  `(c=1 f=VT_20051027-1400)
---------------------------------------------
 Speaker type confusion matrix -- speaker weighted
  REF\SYS (count)      unknown               MISS              
unknown                   2 / 100.0%          0 /   0.0%
  FALSE ALARM             0 /   0.0%
---------------------------------------------
 Speaker type confusion matrix -- time weighted
  REF\SYS (seconds)    unknown               MISS              
unknown                4.65 /  68.4%       2.15 /  31.6%
  FALSE ALARM          0.25 /   3.7%
---------------------------------------------

*** Performance analysis for Speaker Diarization for ALL ***

    EVAL TIME =     37.50 secs
  EVAL SPEECH =     37.50 secs (100.0 percent of evaluated time)
  SCORED TIME =     33.90 secs ( 90.4 percent of evaluated time)
SCORED SPEECH =     33.90 secs (100.0 percent of scored time)
   EVAL WORDS =     23        
 SCORED WORDS =     21         ( 91.3 percent of evaluated words)
---------------------------------------------
MISSED SPEECH =      0.75 secs (  2.2 percent of scored time)
FALARM SPEECH =      0.00 secs (  0.0 percent of scored time)
 MISSED WORDS =      0         (  0.0 percent of scored words)
---------------------------------------------
SCORED SPEAKER TIME =     35.30 secs (104.1 percent of scored speech)
MISSED SPEAKER TIME =      2.15 secs (  6.1 percent of scored speaker time)
FALARM SPEAKER TIME =      0.25 secs (  0.7 percent of scored speaker time)
 SPEAKER ERROR TIME =      0.00 secs (  0.0 percent of scored speaker time)
SPEAKER ERROR WORDS =      3         ( 14.3 percent of scored speaker words)
---------------------------------------------
 OVERALL SPEAKER DIARIZATION ERROR = 6.80 percent of scored speaker time  `(ALL)
---------------------------------------------
 Speaker type confusion matrix -- speaker weighted
  REF\SYS (count)      adult_male          unknown               MISS              
adult_male                1 /  33.3%          0 /   0.0%          0 /   0.0%
unknown                   0 /   0.0%          2 /  66.7%          0 /   0.0%
  FALSE ALARM             0 /   0.0%          0 /   0.0%
---------------------------------------------
 Speaker type confusion matrix -- time weighted
  REF\SYS (seconds)    adult_male          unknown               MISS              
adult_male            28.50 /  80.7%       0.00 /   0.0%       0.00 /   0.0%
unknown                0.00 /   0.0%       4.65 /  13.2%       2.15 /   6.1%
  FALSE ALARM          0.00 /   0.0%       0.25 /   0.7%
---------------------------------------------