AmTrain

The AmTrain task presents training audio and transcription data to the acoustic model training process to create accumulator files. The AmTrainFinal task uses these accumulator files to produce a final adapted acoustic model.

Parameters

Parameter Description Required
Type The task name. Set to AmTrain. Yes
AdaptSil Whether to adapt the silence model.  
Am The acoustic model to adapt. Yes
BeamStep The amount to increase the beam value by on a pass failure, before attempting another pass.  
DataList A list of the adaptation files. Yes
Diag Whether to generate diagnostic information.  
DiagFile The file to write the diagnostic information to.  
Junk Whether to identify words in the adaptation data with poor alignment scores as junk.  
JunkThresh The alignment score threshold. Word alignments scoring above this value are labeled as junk.  
MaxBeam The maximum beam value at which to attempt the adaptation pass.  
MinBeam The minimum beam value at which to attempt the adaptation pass.  
MLLRMaxMins Whether to use standard acoustic adaptation or rapid adaptation mode.  
MLLRMinOcc When the AmTrain task runs in rapid adaptation mode, the minimum number of times that a basic phoneme (for example, ‘d’) must occur in the adaptation data before an individual phoneme transform can be used for adaptation.  
Out The name of the adaptation accumulator (.acc) file to produce. Yes
OutLabExt The label file extension.  
OutLabPath The directory to write label files to. By default, HPE IDOL Speech Server writes the files to the configured temp directory.  
Pgf The pronunciation generation (.pgf) file included in the language pack. Yes
PlhExt The file extension of the input audio feature files.  
PlhPath The path to the directory containing the acoustic feature (.plh) files specified in the DataList. Yes
RelaxRestrain Relaxes time restraints by a specified number of frames.  
Restrain Whether to apply time constraints to the locations of the words in audio during processing.  
SilRestrain Whether to apply time constraints to the locations of silence in audio during processing.  
TxtExt The file extension of the input transcription files.  
TxtPath The path to the directory containing the transcript (.ctm) files specified in the DataList. Yes
WriteOutLabs Whether to create label files.  
ZeroDurWords Whether to label zero-duration words as junk.  

Example

http://localhost:13000/action=AddTask&Type=AmTrain&Am=C:\LP\ENUK\ver-ENUK-5.0-16k.am&Pgf=C:\LP\ENUK\ver-ENUK-5.0.pgf&DataList=ListManager/OptList&PlhPath=C:\data\PLH&TxtPath=C:\data\transcripts&Out=AmAcc.acc 

This action uses port 13000 to instruct HPE IDOL Speech Server, which is located on the local machine, to produce the AmAcc accumulator file using the ver-ENUK-5.0-16k acoustic model, ver-ENUK-5.0 pronunciation generation file, audio feature files stored in C:\data\PLH, and transcription files stored in C:\data\transcripts.


_HP_HTML5_bannerTitle.htm