| 0:00:15 | my little posters about forensic voice comparison |
|---|
| 0:00:21 | and it's |
|---|
| 0:00:23 | takes some examples from real casework to describe |
|---|
| 0:00:29 | a small experiment to find out |
|---|
| 0:00:32 | what might be a slightly better way of doing things because as i've written here |
|---|
| 0:00:37 | when |
|---|
| 0:00:37 | when we do |
|---|
| 0:00:38 | real world forensic voice comparison |
|---|
| 0:00:41 | we want to know what the best approaches to use on the circumstances of the |
|---|
| 0:00:46 | case |
|---|
| 0:00:48 | and this is really the |
|---|
| 0:00:52 | main reason for doing this kind of research simply to find out when you've got |
|---|
| 0:00:57 | of the case the that the case in front of you |
|---|
| 0:01:01 | what's the best |
|---|
| 0:01:02 | approach to use |
|---|
| 0:01:03 | so |
|---|
| 0:01:05 | it's deals with the situation way you have recurrent probably syllabic words like alright crops |
|---|
| 0:01:13 | up a lot not too bad crops up a lot hello okay |
|---|
| 0:01:17 | in both suspect and offender speech samples |
|---|
| 0:01:19 | and |
|---|
| 0:01:21 | if you are using semi |
|---|
| 0:01:23 | semi automatic forensic |
|---|
| 0:01:26 | speaker recognition methods then a one of the main things that is the model they |
|---|
| 0:01:30 | the performance |
|---|
| 0:01:33 | trajectories on the separate syllables like and i |
|---|
| 0:01:36 | all |
|---|
| 0:01:39 | but what i wanted to do was to find out whether you get better strength |
|---|
| 0:01:43 | of evidence |
|---|
| 0:01:44 | if you don't do that but measure of the formant trajectories over the whole of |
|---|
| 0:01:49 | the work |
|---|
| 0:01:50 | a sort of a kind of so to perform well that it from multiple |
|---|
| 0:01:54 | so that this was tested with some high-level features |
|---|
| 0:01:59 | from thee formant a pattern and the tonal fundamental frequency in the cantonese would die |
|---|
| 0:02:05 | which means first |
|---|
| 0:02:08 | and validated we'd likelihood ratios |
|---|
| 0:02:11 | so i obviously there's absolutely nothing whatsoever i can |
|---|
| 0:02:16 | side you about automatic speaker recognition i realise that but they might be some |
|---|
| 0:02:20 | interesting aspects |
|---|
| 0:02:23 | concerning the simple a high-level will to the that we that we working and i |
|---|
| 0:02:29 | do have some interesting things to say that likelihood ratios which you might want to |
|---|
| 0:02:33 | come check with me about and |
|---|
| 0:02:37 | the main reason on here goes is to is to how you can help me |
|---|
| 0:02:41 | out so please come have a chat thank you |
|---|