The SpkIdEvalStream
task is deprecated in IDOL Speech Server version 11.4 and later. Use the IvSpkId task instead.
This task is still available for existing implementations, but it might be incompatible with new functionality. The task might be deleted in future.
After you have trained a set of speaker templates, use the SpkIdEvalStream
task to identify any sections of an audio stream where the trained speakers are present.
To process an audio file, use the SpkIdEvalWav task.
Parameter | Description | Required |
---|---|---|
Type | The task name. Set to SpkIdEvalStream . |
Yes |
AllowEmpty | Whether to produce gender labels as output if no speakers are specified. | |
ClosedSet | Whether the task that you are running is a closed-set test. | |
DiagFile | The name of the file to write diagnostic information to. | |
DiagLevel | The level of detail to include in the diagnostic information. | |
DiscardShort | Exclude segments shorter than a specific duration from further analysis. | |
MinNonSpeech | The minimum length of non-speech segments. | |
MinSpeech | The minimum length of speech segments. | |
Out | The file to write the results to. | |
Sfreq | The sample frequency of the audio stream to process. | |
The file extension to use for template files. | ||
TemplateList | A list file that lists multiple speaker template files to use. | Yes, if TemplateSet is not specified |
TemplatePath | The path to the directory containing the speaker templates. | |
TemplateSet | An audio template set file. | Yes, if TemplateList is not specified |
ThreshScale | The rate at which to scale the thresholds. |
http://localhost:15000/action=AddTask&Type=SpkIdEvalStream&File=C:\Data\Speech.wav&TemplateSet=speakers.ats&ClosedSet=False&Out=results.ctm
This action uses port 15000
to instruct IDOL Speech Server, which is located on the local machine, to search the audio stream for speakers based on the template set file speakers.ats
, and to write the identification results to the results.ctm
file. Because the test is set to be open-set, IDOL Speech Server marks sections where no speaker scores above their respective thresholds as Unknown_
.
|