0:00:16 | everybody i in a single button so that a new feature for automatic speaker verification |
---|
0:00:22 | spoofing |
---|
0:00:26 | the data for us to show that automatic specific speaker verification system |
---|
0:00:33 | are all the spoofing |
---|
0:00:35 | so |
---|
0:00:38 | especially when no we are in an ideal scenario where the spoofing we cannot be |
---|
0:00:44 | you know in advance so the need the all the for the generalize the going |
---|
0:00:50 | to measure |
---|
0:00:51 | so we propose an feature example sample transform |
---|
0:00:56 | and the t v is a |
---|
0:00:59 | task automation |
---|
0:01:02 | and different from a for their trust one because we suppose that fourier transform me |
---|
0:01:06 | like frequency resolution secrecy security uses our body a time frequency the solution that means |
---|
0:01:14 | i your time resolution for i frequencies and the higher frequency resolution for lower frequencies |
---|
0:01:22 | we combine the c d more or less with the traditional a cepstral analysis but |
---|
0:01:30 | we found the problem but using the discrete cosine transform or you know the identity |
---|
0:01:37 | applied before so |
---|
0:01:39 | for two reasons results secrecy a dct have different skate one dramatic another one is |
---|
0:01:46 | the dinner and dramatically dct basis are no longer than one |
---|
0:01:50 | so we found a total shock joe's the only for me to sing but they're |
---|
0:01:54 | not but for thirty twenty scale over the speaker i |
---|
0:01:58 | with a linear frequency scale |
---|
0:02:00 | a |
---|
0:02:02 | these is the some form comparison of a results a nice peaceful for database and |
---|
0:02:08 | we know that |
---|
0:02:13 | we found that no i x from a tax or the system and earlier are |
---|
0:02:18 | excellent error rate for unknown i so the thing to me |
---|
0:02:25 | best performance especially for a s then that young federation seem to use these your |
---|
0:02:31 | voice |
---|
0:02:32 | and where a we obtain a seven percent over relative improvement |
---|
0:02:38 | embodied the exposing detection performance |
---|
0:02:43 | a part of to be in our work seeks |
---|
0:02:47 | we that seventy two relative improvement |
---|
0:02:49 | so for more details are wait for you |
---|
0:02:54 | sure some possible |
---|