NIST-SRE04-16 Dataset¶
Dataset Description¶
This is an aggregation of the NIST-SRE datasets from 2004 to 2016.
Identities |
Sample count |
||
train |
6213 |
71728 |
|
dev |
references |
80 |
120 |
probes |
5 |
1207 |
|
eval |
references |
802 |
1202 |
probes |
5 |
9294 |
GMM¶
Development |
Evaluation |
Command used:
$ bob bio pipeline -d nist-sre04to16 -p gmm-nist -g dev -g eval -l sge -o results/gmm_nist
On 1281 CPU nodes on the SGE Grid: TODO
ISV¶
Development |
Evaluation |
Command used:
$ bob bio pipeline -d nist-sre04to16 -p isv-nist -g dev -g eval -l sge -o results/isv_nist
On 1281 CPU nodes on the SGE Grid: TODO
Speechbrain ECAPA-TDNN¶
Development |
Evaluation |
|
---|---|---|
Failure to Acquire |
0.0% |
0.0% |
False Match Rate |
28.4% (27400/96342) |
28.7% (2142253/7453619) |
False Non Match Rate |
28.4% (62/218) |
33.7% (57/169) |
False Accept Rate |
28.4% |
28.7% |
False Reject Rate |
28.4% |
33.7% |
Half Total Error Rate |
28.4% |
31.2% |
Command used:
$ bob bio pipeline -d nist-sre04to16 -p speechbrain-ecapa-voxceleb -g dev -g eval -l sge -o results/speechbrain_nist
On 1281 CPU nodes on the SGE Grid: Ran in 50 minutes (no training).
Footnotes