(Applies to the IDOL Speech Server component.)
To perform speaker identification and segmentation in IDOL Speech Server, you must create and optimize speaker templates for each speaker that you want to identify, and create speaker classifier templates. The Create Speaker Classifier wizard in the Control menu takes you through this process.
The AddTask
and WavSidOptimize
actions perform these tasks. For more information on speaker templates, see the IDOL Speech Server Administration Guide and the IDOL Speech Server Reference.
To create a speaker classifier template
Click Choose file and browse to the location of an audio file containing sample speech from the person you chose in step 7.
Do not use the same audio that was used to create the speaker templates.
|