IDOL Speech Server initially segments audio by speaker and then identifies the speaker in each segment. Short segments can lower identification accuracy and you can choose to exclude short segments from speaker identification by setting DiscardShortSegs
to a minimum segment length.
speakerid
module returns a result of 'Unknown
' for any section with a duration below the specified value.audiotemplatescore
and ivscore
modules return a result of 'Unknown
' for any section with a duration below the specified value.
Type: | Decimal |
Default: | speakerid module: 1.0audiotemplatescore module: 1.0ivscore module: 2.0 |
Required: | No |
Configuration Section: | audiotemplatescore, ivscore, and speakerid modules |
Example: | DiscardShortSegs=3.0
|
See Also: | DiscardShort (action parameter) MinNonSpeech MinSpeech |
|