FixTime

A fixed time duration for speaker clustering. For example, if you are processing a media file that is 30 minutes in length you can crystallize the speaker information after five minutes. Any subsequent speech segments are then assigned to one of the speakers from the first five minutes of the file (Media Server does not consider that they could represent new speakers). This increases the speed of classification for the rest of the file.

To switch off this functionality, set this parameter to 0.

Type:
Default: 8 minutes
Required: No
Configuration Section: TaskName
Example: FixTime=5 minutes
See Also: