The frame normalization to use.
Set FrameNorm
to one of the following options:
NONE
UBM
SET
AUTO
(audiotemplatescore
module only)The frame normalization used during the identification stage must match that used during development.
By default, HPE IDOL Speech Server uses UBM
if the templates are using a background model.
This means that when HPE IDOL Speech Server generates scores for each audio frame, the server takes the background model score into account and subtracts the background model score from all the speaker scores. This process helps to reduce frame score variation (in cases where some frames might score higher on all models, including the background, leading to unstable results). HPE recommends that you use the UBM
option for open-set speaker identification.
If you set FrameNorm
to NONE
, HPE IDOL Speech Server does not take the background model score into account, but instead uses the scores as they are. HPE recommends that you use this option only if you trained your templates without using a Universal Background Model.
If you set FrameNorm
to SET
, HPE IDOL Speech Server takes the scores for all templates, and normalizes them between 0.0 (the worst case) and 1.0 (the best case). This process can increase recall, but can also introduce more false hits. HPE recommends that you use the SET
option for closed-set speaker identification.
If you set FrameNorm
to AUTO
in the audiotemplatescore
module, HPE IDOL Speech Server uses the frame normalization that was used during development when you produced the scores for threshold estimation. If development was not performed and there are no thresholds, HPE IDOL Speech Server switches to UBM
mode, unless you set ClosedSet
to True
, in which case HPE IDOL Speech Server uses SET
mode.
Type: | String |
Default: | UBM in audiotemplatedevel moduleAUTO in audiotemplatescore module
|
Required: | No |
Configuration Section: | audiotemplatedevel module |
Example: | FrameNorm=SET
|
See Also: |
FrameNorm (action parameter) |
|