command line (run on 2004 Oct 29 at 14:17:53): ../src/md-eval-v19a.pl -af -e -D -d -W -w -t 1.0 -l 2 -u md_test23.uem -r md_test23.ref.rttm -s md_test23.sys.rttm Word-based metadata alignment, max gap between matching words = 1.0 sec Metadata evaluation parameters: word-optimized metadata mapping max gap between matching metadata events = 0.1 words max extent to match for SU's = 2 words 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 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/ X X X X NO_RT_METADATA/ X SU/unannotated X Word alignment and scoring details for channel 1 of file md_test23 ref del ins sub REF: token type tbeg tend speaker SYS: token type Rtbeg Rtend tbeg tend sys-speaker 1 - - 0 firstword lex 10.00 11.00 spkr1r firstword lex ( 10.00 11.00) 10.00 11.00 spkr1s 1 - - 0 commercial lex 11.00 12.00 spkr1r commercial lex ( 11.00 12.00) 11.00 12.00 spkr1s 1 - - 0 message lex 12.00 13.00 spkr1r message lex ( 12.00 13.00) 12.00 13.00 spkr1s 1 - - 0 fourthword lex 13.00 13.50 spkr1r fourthword lex ( 13.00 13.50) 13.00 13.50 spkr1s 1 - - 0 fifthword lex 13.50 14.00 spkr1r fifthword lex ( 13.50 14.00) 13.50 14.00 spkr1s 1 - - 0 sixthword lex 14.00 15.00 spkr2r sixthword lex ( 14.00 15.00) 14.00 15.00 spkr2s 1 - - 0 seventhword lex 15.00 16.00 spkr2r seventhword lex ( 15.00 16.00) 15.00 16.00 spkr2s 1 - - 0 eighthword lex 17.50 18.00 spkr2r eighthword lex ( 17.50 18.00) 17.50 18.00 spkr2s 1 - - 0 ninthword lex 18.00 19.00 spkr2r ninthword lex ( 18.00 19.00) 18.00 19.00 spkr2s 1 - - 0 tenthword lex 19.00 20.00 spkr2r tenthword lex ( 19.00 20.00) 19.00 20.00 spkr2s 1 - - 0 eleventhword lex 20.00 21.00 spkr2r eleventhword lex ( 20.00 21.00) 20.00 21.00 spkr2s 1 - - 0 twelfthword lex 21.00 22.00 spkr2r twelfthword lex ( 21.00 22.00) 21.00 22.00 spkr2s 1 - - 0 thirteenthword lex 22.00 23.00 spkr2r thirteenthword lex ( 22.00 23.00) 22.00 23.00 spkr2s SU alignment and scoring details for channel 1 of file md_test23 ref del ins sub REF: token type tbeg tend speaker SYS: token type Rtbeg Rtend tbeg tend sys-speaker 1 - - 0 statement SU 10.00 11.00 spkr1r statement SU ( 10.00 11.00) 10.00 11.00 spkr1s 1 - - 0 statement SU 13.00 14.00 spkr1r statement SU ( 13.00 14.00) 13.00 14.00 spkr1s 1 - - 0 statement SU 14.00 15.00 spkr2r statement SU ( 14.00 15.00) 14.00 15.00 spkr2s 1 - - 0 question SU 15.00 18.00 spkr2r question SU ( 15.00 18.00) 15.00 18.00 spkr2s 1 - - 0 statement SU 18.00 20.00 spkr2r statement SU ( 18.00 20.00) 18.00 20.00 spkr2s Chronological display of sys data aligned with ref data for file 'md_test23', channel '1' ----------------------- reference ----------------------- | mapped | --------------------- system output --------------------- --type-- -subtyp- -----word/spkr----- -tbeg- -tend- | ref_ID | --type-- -subtyp- -----word/spkr----- -tbeg- -tend- beg SEGMENT spkr1r 10.00 | | beg SPEAKER child spkr1r 10.00 | | 10.00 | |beg SPEAKER child spkr1s=>spkr1r 10.00 beg SU statemen spkr1r 10.00 | SU1 |beg SU statemen spkr1s=>spkr1r 10.00 LEXEME lex FIRSTWORD 10.00 11.00 | LX1 | LEXEME lex FIRSTWORD 10.00 11.00 end SU statemen spkr1r 11.00 | SU1 |end SU statemen spkr1s=>spkr1r 11.00 end SPEAKER child spkr1r 11.00 | | beg SPEAKER child spkr1r 11.00 | | LEXEME lex COMMERCIAL 11.00 12.00 | LX2 | LEXEME lex COMMERCIAL 11.00 12.00 LEXEME lex MESSAGE 12.00 13.00 | LX3 | LEXEME lex MESSAGE 12.00 13.00 end SPEAKER child spkr1r 13.00 | | beg SPEAKER child spkr1r 13.00 | | beg SU statemen spkr1r 13.00 | SU2 |beg SU statemen spkr1s=>spkr1r 13.00 LEXEME lex FOURTHWORD 13.00 13.50 | LX4 | LEXEME lex FOURTHWORD 13.00 13.50 LEXEME lex FIFTHWORD 13.50 14.00 | LX5 | LEXEME lex FIFTHWORD 13.50 14.00 end SU statemen spkr1r 14.00 | SU2 |end SU statemen spkr1s=>spkr1r 14.00 end SPEAKER child spkr1r 14.00 | | 14.00 | |end SPEAKER child spkr1s=>spkr1r 14.00 end SEGMENT spkr1r 14.00 | | beg SEGMENT spkr2r 14.00 | | beg SPEAKER adult_ma spkr2r 14.00 | | 14.00 | |beg SPEAKER adult_ma spkr2s=>spkr2r 14.00 beg SU statemen spkr2r 14.00 | SU3 |beg SU statemen spkr2s=>spkr2r 14.00 LEXEME lex SIXTHWORD 14.00 15.00 | LX6 | LEXEME lex SIXTHWORD 14.00 15.00 end SU statemen spkr2r 15.00 | SU3 |end SU statemen spkr2s=>spkr2r 15.00 beg SU question spkr2r 15.00 | SU4 |beg SU question spkr2s=>spkr2r 15.00 LEXEME lex SEVENTHWORD 15.00 16.00 | LX7 | LEXEME lex SEVENTHWORD 15.00 16.00 NON-LEX sneeze AHHCHOO 16.00 17.50 | | LEXEME lex EIGHTHWORD 17.50 18.00 | LX8 | LEXEME lex EIGHTHWORD 17.50 18.00 end SU question spkr2r 18.00 | SU4 |end SU question spkr2s=>spkr2r 18.00 beg SU statemen spkr2r 18.00 | SU5 |beg SU statemen spkr2s=>spkr2r 18.00 LEXEME lex NINTHWORD 18.00 19.00 | LX9 | LEXEME lex NINTHWORD 18.00 19.00 LEXEME lex TENTHWORD 19.00 20.00 | LX10 | LEXEME lex TENTHWORD 19.00 20.00 end SU statemen spkr2r 20.00 | SU5 |end SU statemen spkr2s=>spkr2r 20.00 end SPEAKER adult_ma spkr2r 20.00 | | end SEGMENT spkr2r 20.00 | | beg SPEAKER adult_ma spkr2r 20.00 | | LEXEME lex ELEVENTHWORD 20.00 21.00 | LX11 | LEXEME lex ELEVENTHWORD 20.00 21.00 LEXEME lex TWELFTHWORD 21.00 22.00 | LX12 | LEXEME lex TWELFTHWORD 21.00 22.00 LEXEME lex THIRTEENTHWORD 22.00 23.00 | LX13 | LEXEME lex THIRTEENTHWORD 22.00 23.00 end SPEAKER adult_ma spkr2r 23.00 | | 23.00 | |end SPEAKER adult_ma spkr2s=>spkr2r 23.00 *** Performance analysis for SUs *** overall error SCORE = 0.00% SU (exact) end detection statistics -- in terms of reference words Nref Ndel Nins Nsub %Del %Ins %Sub %D+I %Tot ALL 5 0 0 0 0.00 0.00 0.00 0.00 0.00 SU detection statistics -- in terms of # of SUs Nref Ndel Nins Nsub %Del %Ins %Sub %D+I %Tot ALL 5 0 0 0 0.00 0.00 0.00 0.00 0.00 f=md_test23 5 0 0 0 0.00 0.00 0.00 0.00 0.00 SU detection confusion matrix -- in terms of # of SUs ALL - ref\sys question statemen {Miss} question 1 0 0 statement 0 4 0 {FA} 0 0 SU word offset statistics for ALL data word offsets: <-3 -3 -2 -1 0 1 2 3 >3 BEG: 0 - - - 5 - - - 0 END: 0 - - - 5 - - - 0 *** Performance analysis for Speaker Diarization for f=md_test23 *** EVAL TIME = 9.00 secs EVAL SPEECH = 9.00 secs (100.0 percent of evaluated time) SCORED TIME = 7.50 secs ( 83.3 percent of evaluated time) SCORED SPEECH = 7.50 secs (100.0 percent of scored time) EVAL WORDS = 9 SCORED WORDS = 9 (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 = 7.50 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 `(f=md_test23) --------------------------------------------- Speaker type confusion matrix -- speaker weighted REF\SYS (count) adult_male child MISS adult_male 1 / 50.0% 0 / 0.0% 0 / 0.0% child 0 / 0.0% 1 / 50.0% 0 / 0.0% FALSE ALARM 0 / 0.0% 0 / 0.0% --------------------------------------------- Speaker type confusion matrix -- time weighted REF\SYS (seconds) adult_male child MISS adult_male 5.50 / 73.3% 0.00 / 0.0% 0.00 / 0.0% child 0.00 / 0.0% 2.00 / 26.7% 0.00 / 0.0% FALSE ALARM 0.00 / 0.0% 0.00 / 0.0% --------------------------------------------- *** Performance analysis for Speaker Diarization for ALL *** EVAL TIME = 9.00 secs EVAL SPEECH = 9.00 secs (100.0 percent of evaluated time) SCORED TIME = 7.50 secs ( 83.3 percent of evaluated time) SCORED SPEECH = 7.50 secs (100.0 percent of scored time) EVAL WORDS = 9 SCORED WORDS = 9 (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 = 7.50 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 `(ALL) --------------------------------------------- Speaker type confusion matrix -- speaker weighted REF\SYS (count) adult_male child MISS adult_male 1 / 50.0% 0 / 0.0% 0 / 0.0% child 0 / 0.0% 1 / 50.0% 0 / 0.0% FALSE ALARM 0 / 0.0% 0 / 0.0% --------------------------------------------- Speaker type confusion matrix -- time weighted REF\SYS (seconds) adult_male child MISS adult_male 5.50 / 73.3% 0.00 / 0.0% 0.00 / 0.0% child 0.00 / 0.0% 2.00 / 26.7% 0.00 / 0.0% FALSE ALARM 0.00 / 0.0% 0.00 / 0.0% ---------------------------------------------