The LangIdBndStream
task receives audio data as a binary stream, converts it into language identification features, and determines boundaries where the language changes. The task returns the language identification results between boundaries.
Parameter | Description | Required |
---|---|---|
Type | The task name. Set to LangIdBndStream . |
Yes |
AppDnnBase | The location of the appResources directory, which contains the DNN and .ian files to use. |
|
Beam | The beam width of the search process. | |
ClassList | A list of language classifiers to use. | |
ClassPath | The path to the directory containing the language classifiers. | |
DnnFile | The Deep Neural Network acoustic modeling file to use. | |
FrameDupl | The balance between performance and speed for audio preprocessing DNN classification. | |
Lang | The name of the language pack to use. | Yes |
LangList | A subset of languages to use from the classifier list file. | |
MinPhoneRate | The minimum phone rate (phones per second). | |
Mode | The algorithm mode for the speech-to-text process. | |
NBest | The maximum number of language candidates to include in the output file. | |
Out | The file to write language identification results to. | Yes |
OutB | The file to write boundary point information to. | Yes |
SegSize | The maximum results segment size. | |
SegSmoothWin | The size of the smoothing window. | |
SegStep | The step size in phones of the analysis window. | |
SilThresh | The threshold between what the task identifies as silence and non-silence. | |
SpeechThresh | The threshold between speech and non-speech (music or noise). |
Note: The ClassList
parameter is required only if you want to change the audio sample rate, or if you want to use your own custom classifiers. You might also need to specify the ClassPath
parameter, depending on the location of the classifier files.
http://localhost:13000/action=AddTask&Type=LangIdBndStream&ClassList=ListManager/OptClassSet&ClassPath=C:\LangID\&Out=SpeechLang1.ctm&OutB=SpeechBnd.ctm
This action uses port 13000
to instruct HPE IDOL Speech Server, which is located on the local machine, to identify languages and language boundaries in the audio stream using the language classifiers specified in the OptClassSet
list. The action instructs HPE IDOL Speech Server to write the identification results to the SpeechLang1.ctm
file and the boundary information to the SpeechBnd.ctm
file.
|