The sidtrain
module trains HPE IDOL Speech Server for speaker identification.
Mode | Input | Output | Description |
---|---|---|---|
f , w |
Accepts feature frames, and trains the speaker model. Also accepts filter labels to determine which areas to process as speech. |
Example:
output <- sidtrain(_, f, w)
AdaptRel * | NMix |
Diag | OrgModel* |
DiagFile | SampleFrequency |
DiagFileLabel | SpkIDBasePack* |
MinFrames | USMEnabled * |
NewModel | USMFile * |
NewModelLabel* |
*You must set the value of these configuration parameters by setting the corresponding action parameters.
|