![Background
model
training
IteraBve
128
Gaussian
Spling
EM-‐ML
GMM
training
K-‐means
assignment
[1]
“Speaker
Independent
discriminant
feature
extracBon
for
acousBc
paXern
matching”,
Xavier
Anguera,
ICASSP
2012](https://image.slidesharecdn.com/tidmediaeval2012-121005120456-phpapp02/85/Telefonica-Research-System-for-the-Spoken-Web-Search-task-at-Mediaeval-2012-5-320.jpg)
![DTW
local
constraints
• no
global
constraints
are
applied
in
order
to
allow
for
matching
of
any
segment
among
both
sequences
• Local
constraints
are
set
to
allow
warping
up
to
2X
" D(m ! 2, n) + d(xm , yn ) (m,
n)
$
$ jumps(m ! 2, n) + 3
$ D(m, n ! 2) + d(xm , yn ) (m-‐2,
n-‐1)
D(m, n) = min #
$ jumps(m, n ! 2) + 3
$ D(m ! 2, n ! 2) + d(x , y )
m n
$ (m-‐1,
n-‐2)
% jumps(m ! 2, n ! 2) + 4 (m-‐1,
n-‐1)
• Posteriorgram
features
distance:
$ N!1 '
d(xm , yn ) = ! log & # xm [i]" yn [i])
% i=0 (](https://image.slidesharecdn.com/tidmediaeval2012-121005120456-phpapp02/85/Telefonica-Research-System-for-the-Spoken-Web-Search-task-at-Mediaeval-2012-11-320.jpg)
Telefonica Research System for the Spoken Web Search task at Mediaeval 2012
- 1. Telefonica Research at Mediaeval 2012 Spoken Web Search Task Xavier Anguera
- 2. Outline • System descripBon – Speech AcBvity detecBon • Proposed systems – Segmental-‐DTW – IR-‐DTW • Results
- 3. Proposed overall system S-‐DTW IR-‐DTW
- 4. Frontend MFCC-‐39 features (12 Cepstra + Energy) + Delta + DeltaDelta Mean & variance normalizaBon at sentence level Posterior probabiliBes from a GMM background model L2-‐normalizaBon
- 5. Background model training IteraBve 128 Gaussian Spling EM-‐ML GMM training K-‐means assignment [1] “Speaker Independent discriminant feature extracBon for acousBc paXern matching”, Xavier Anguera, ICASSP 2012
- 6. Silence modeling 10% lowest energy frames • 1 Gauss for noise and 4 Gauss for speech Silence/Speech • Perform 10 iteraBons or GMM training while % variaBon is high Decode the data
- 7. 2234444343322444444444443222222234444444444444444444444443210000011222443 Threshold set to values <2 (i.e. silence + lowest speech)
- 8. Overlap postprocessing • We compute the percentage of overlap between all matching paths min(End1, End2) ! max(Start1, Start2) Ovl = min(End1! Start1, End2 ! Start2) • For pairs with > 0.5 overlap – Select the match with highest score
- 9. Start1 End1 Match1 Match2 Start2 End2 min(ends) – max(starts) Ovl = = 0.8 Min(size1, size2)
- 10. S-‐DTW submission • Based on last year’s submission but with the system improvements above
- 11. DTW local constraints • no global constraints are applied in order to allow for matching of any segment among both sequences • Local constraints are set to allow warping up to 2X " D(m ! 2, n) + d(xm , yn ) (m, n) $ $ jumps(m ! 2, n) + 3 $ D(m, n ! 2) + d(xm , yn ) (m-‐2, n-‐1) D(m, n) = min # $ jumps(m, n ! 2) + 3 $ D(m ! 2, n ! 2) + d(x , y ) m n $ (m-‐1, n-‐2) % jumps(m ! 2, n ! 2) + 4 (m-‐1, n-‐1) • Posteriorgram features distance: $ N!1 ' d(xm , yn ) = ! log & # xm [i]" yn [i]) % i=0 (
- 12. S-‐DTW algorithm Query term Reference term
- 13. S-‐DTW algorithm Query term Reference term
- 14. IR-‐DTW • Total rework from last year’s system • Aim at keeping the same accuracy, but: – Much less memory usage – Faster retrieval • IR (InformaBon Retrieval) cause we use reference features indexing for fast nearest neighbors retrieval
- 15. Official results MTWV Dev-‐dev Dev-‐eval Eval-‐dev Eval-‐eval IR-‐DTW 0.3903 0.3139 0.4983 0.3416 S-‐DTW 0.3745 0.3001 0.4716 0.3113 ATWV Dev-‐dev Dev-‐eval Eval-‐dev Eval-‐eval IR-‐DTW 0.3866 0.3042 0.4219 0.3301 S-‐DTW 0.3644 0.292 0.3988 0.2942
- 16. DEV-DEV results 98 Random Performance IR-DTW MTWV=0.390 Scr=0.387 95 S-DTW MTWV=0.375 Scr=0.695 90 80 Miss probability (in %) 60 40 20 10 5 .0001 .001 .004 .01.02 .05 .1 .2 .5 1 2 5 10 20 40 False Alarm probability (in %)
- 17. EVAL-EVAL Results 98 Random Performance IR-DTW MTWV=0.342 95 S-DTW MTWV=0.311 90 80 Miss probability (in %) 60 40 20 10 5 .0001 .001 .004 .01.02 .05 .1 .2 .5 1 2 5 10 20 40 False Alarm probability (in %)
- 18. DEV-EVAL results 98 Random Performance IR-DTW MTWV=0.314 95 S-DTW MTWV=0.300 90 80 Miss probability (in %) 60 40 20 10 5 .0001 .001 .004 .01.02 .05 .1 .2 .5 1 2 5 10 20 40 False Alarm probability (in %)
- 19. EVAL-DEV results 98 Random Performance IR-DTW MTWV=0.498 95 S-DTW MTWV=0.472 90 80 Miss probability (in %) 60 40 20 10 5 .0001 .001 .004 .01.02 .05 .1 .2 .5 1 2 5 10 20 40 False Alarm probability (in %)
- 20. Xavier Anguera Summary xanguera@Bd.es • We propose 2 systems, all sharing the same framework • Some improvements in the framework were incorporated: speech/silence classificaBon, new overlap detecBon, modified background model. • IR-‐DTW is a total reimplementaBon of SDTW, using informaBon retrieval concepts