| Class | Description |
|---|---|
| AbstractTimelineMaker | |
| AcousticFeatureFileWriter | |
| AllophonesExtractor |
For the given texts, compute allophones, especially boundary tags.
|
| AutocorrelationPitchmarker |
A pitch marker component that uses F0TrackerAutocorrelationHeuristic.
|
| BasenameTimelineMaker |
The BasenameTimelineMaker class takes a database root directory and a list of basenames, and associates the basenames with
absolute times in a timeline in Mary format.
|
| CARTBuilder | |
| CARTBuilder.WagonCallerThread | |
| DatabaseImportMain |
The single purpose of the DatabaseImportMain class is to provide a main which executes the sequence of database import and
conversion operations.
|
| DatabaseLayout |
The DatabaseLayout class registers the base directory of a voice database, as well as the various subdirectories where the
various voice database components should be stored or read from.
|
| DurationCARTTrainer |
A class which converts a text file in festvox format into a one-file-per-utterance format in a given directory.
|
| DurationSoPTrainer |
Modelling duration using Sum of products (SoP) SoP is modelled using multiple linear regression Selection of features is
performed with Sequential Floating Forward Search(SFFS):
|
| DurationTreeTrainer |
A class which converts a text file in festvox format into a one-file-per-utterance format in a given directory.
|
| EHMMLabeler |
Automatic Labelling using EHMM labeller
|
| EndpointDetector |
Identify and Remove End-ponints (intitial and final silences) from given set of wave files.
|
| ESTCaller |
The ESTCaller class emulates the behaviour of the EST-calling shell scripts for the calculation of pitchmarks, lpc,
mel-cepstrum etc.
|
| F0CARTTrainer |
A class which converts a text file in festvox format into a one-file-per-utterance format in a given directory.
|
| F0PolynomialFeatureFileWriter | |
| F0PolynomialInspector | |
| F0PolynomialTreeTrainer | |
| F0SoPTrainer |
Example of how to use SoP modules in a voice configuration file:
#Modules to use for predicting acoustic target features for this voice: #this are the names that will be used to identify
diferent models of the same type, sop, cart or hmm voice.cmu-slt-hsmm.acousticModels = duration F0 midF0 rightF0
voice.cmu-slt-hsmm.duration.model = sop voice.cmu-slt-hsmm.duration.data = MARY_BASE/lib/voices/cmu-slt-hsmm/dur.sop
voice.cmu-slt-hsmm.duration.attribute = d
voice.cmu-slt-hsmm.leftF0.model = sop voice.cmu-slt-hsmm.leftF0.data = MARY_BASE/lib/voices/cmu-slt-hsmm/f0.left.sop
voice.cmu-slt-hsmm.leftF0.attribute = f0 voice.cmu-slt-hsmm.leftF0.attribute.format = (0,%.0f)
voice.cmu-slt-hsmm.midF0.model = sop voice.cmu-slt-hsmm.midF0.data = MARY_BASE/lib/voices/cmu-slt-hsmm/f0.mid.sop
voice.cmu-slt-hsmm.midF0.attribute = f0 voice.cmu-slt-hsmm.midF0.attribute.format = (50,%.0f)
voice.cmu-slt-hsmm.rightF0.model = sop voice.cmu-slt-hsmm.rightF0.data = MARY_BASE/lib/voices/cmu-slt-hsmm/f0.right.sop
voice.cmu-slt-hsmm.rightF0.attribute = f0 voice.cmu-slt-hsmm.rightF0.attribute.format = (100,%.0f)
|
| FeatureSelection |
For the given texts, compute unit features and align them with the given unit labels.
|
| FestivalCARTImporter |
Class for importing CARTs from Festival Text-Format to Mary Bin-Format
|
| Festvox2MaryTranscripts |
A class which converts a text file in festvox format into a one-file-per-utterance format in a given directory.
|
| HalfPhoneFeatureFileWriter | |
| HalfPhoneLabelFeatureAligner | |
| HalfPhoneUnitFeatureComputer | |
| HalfPhoneUnitfileWriter | |
| HalfPhoneUnitLabelComputer |
Compute unit labels from phone labels.
|
| HelpGUI | |
| HMMParameterExtractor |
A component for extracting mfccs, lab, state labels, and hmm generated waves (options given to choose)
|
| HMMVoiceCompiler | |
| HMMVoiceConfigure | |
| HMMVoiceDataPreparation |
This program was modified from previous version to: 1.
|
| HMMVoiceFeatureSelection | |
| HMMVoiceMakeData | |
| HMMVoiceMakeVoice | |
| HnmTimelineMaker |
HnmTimelineMaker class takes a database root directory and a list of basenames, and converts the related wav files into a hnm
timeline in Mary format.
|
| HTKLabeler |
Automatic Labelling using HTK labeller
|
| JoinCostFileMaker | |
| JoinCostPrecomputer | |
| JoinModeller | |
| LabelledFilesInspector |
For the given texts, compute unit features and align them with the given unit labels.
|
| LabelPauseDeleter | |
| LPCTimelineMaker |
The LPCTimelineMaker class takes a database root directory and a list of basenames, and converts the related wav files into a
LPC timeline in Mary format.
|
| Mary2FestvoxTranscripts |
Makes the file etc/.txt.done.data from the text files in text directory
|
| MCEPMaker | |
| MCepTimelineMaker |
The mcepTimelineMaker class takes a database root directory and a list of basenames, and converts the related wav files into a
mcep timeline in Mary format.
|
| MRPALabelConverter | |
| NullUnit |
Null-Unit entry for the Catalog.
|
| OctaveVoiceQualityProcessor | |
| PauseDurationTrainer | |
| PauseFeatureComputer |
For the given texts, compute unit features and align them with the given unit labels.
|
| PhoneFeatureFileWriter | |
| PhoneLabelFeatureAligner |
Compare unit label and unit feature files.
|
| PhoneLabelFeatureAligner.EditFrameShower | |
| PhoneUnitFeatureComputer |
For the given texts, compute unit features and align them with the given unit labels.
|
| PhoneUnitfileWriter | |
| PhoneUnitLabelComputer |
Compute unit labels from phone labels.
|
| PraatPitchmarker | |
| QualityControl |
Quality Control Component for Voice Import Tool to perform 'Sensibility check' on Data.
|
| SanityChecker |
Sanity checker for voicebuilding One test case : check no.
|
| SCostUnitFileWriter | |
| SettingsGUI | |
| SnackPitchmarker | |
| SnackVoiceQualityProcessor | |
| SphinxLabeler |
Preparate the directory of the voice for sphinx labelling
|
| SphinxLabelingPreparator |
Preparate the directory of the voice for sphinx labelling
|
| SphinxTrainer |
Class to train sphinx labeler
|
| SPTKMFCCExtractor |
A component for extracting mfcc files from wave files
|
| TimelineWriter |
The TimelineWriter class provides an interface to create or update a Timeline data file in Mary format, and to feed new
datagrams to the timeline file.
|
| TranscriptionAligner | |
| Unit |
Simple Unit entry for the Catalog.
|
| UnitLabel | |
| UnknownWordsFrequencyComputer |
For the given texts, compute unit features and align them with the given unit labels.
|
| VoiceCompiler | |
| VoiceCompiler.MavenVoiceCompiler | |
| VoiceImportComponent |
A component in the process of importing a voice into MARY format.
|
| VoiceQualityTimelineMaker |
Make a Timeline from a directory of voice quality parameter files
|
| WagonCaller |
Class to call Wagon
|
| WaveTimelineMaker |
The WaveTimelineMaker class takes a database root directory and a list of basenames, and split the waveforms as datagrams to be
stored in a timeline in Mary format.
|
| WavReader |
File reader for a wave (RIFF) waveform
|
Copyright © 2000–2016 DFKI GmbH. All rights reserved.