The maximum size (in seconds) of non-speech segments.
You might want to restrict the maximum length of non-speech segments in speaker identification because HPE IDOL Speech Server must wait for a segment to finish before it runs speaker segmentation and identification.
Reducing the maximum segment size might improve the latency in speaker identification. However, reducing it by too much might affect the accuracy of the results.
By default, the maximum length of non-speech segments is not limited.
Action: | AddTask
CheckResources |
Task: | IvSpkIdEvalStream
IvSpkIdEvalWav |
Type: | Decimal |
Default: | 0.0 (unlimited) |
Example: | MaxNonSpeech=5.0
|
See Also: | MaxNonSpeech (configuration parameter) |
|