Shout_Cluster Class Reference

This class is the base class for the shout_cluster application. It segments an audio file based on the speaker clustering procedure from the Train_Speaker_Segmenter class. More...

List of all members.

Public Member Functions
	Shout_Cluster (char *featureFile, char *label, char *inputRTTM_train, char *inputRTTM_decode, char *outputRTTM, char *fea2, double initWeight, char *fea3, int fastMerge, int maxClusters, bool widen)
	~Shout_Cluster ()
Protected Attributes
Adapt_Segmenter *	segmenter
Train_Segmenter *	segmenter

---

Detailed Description

This class is the base class for the shout_cluster application. It segments an audio file based on the speaker clustering procedure from the Train_Speaker_Segmenter class.

---

Constructor & Destructor Documentation

Shout_Cluster::Shout_Cluster	(	char *	featureFile,
		char *	label,
		char *	inputRTTM_train,
		char *	inputRTTM_decode,
		char *	outputRTTM,
		char *	featureFile2,
		double	initWeight,
		char *	featureFile3,
		int	fastMerge,
		int	maxClusters,
		bool	widen
	)

This constructor reads a feature file in order to segment and cluster it.

Energy based Speaker Activity Detection is used for a first rough speech/non-speech clustering. After that, the method in Train_Speaker_Segmenter is used to obtain a speaker segmentation.

References FeaturePool::claimSegmentationID(), FeaturePool::createNewMultiStreamPool(), FeaturePool::createNewPool(), FeaturePool::enableFilter(), FILE_TYPE_RAW_DIARIZATION, FILE_TYPE_SHOUT, FILE_TYPE_TEXT, FeaturePool_Multistream_file::fileName, FeaturePool_Multistream_file::fileType, FeaturePool::getPoolLength(), FeaturePool::readRTTM(), segmenter, and Train_Segmenter::train().

Here is the call graph for this function:

Shout_Cluster::~Shout_Cluster

(

)

The destructor is empty.

---

Member Data Documentation

Train_Segmenter* Shout_Cluster::segmenter [protected]

Adapt_Segmenter* Shout_Cluster::segmenter [protected]

Referenced by Shout_Cluster().