public class HnmVoiceDataDumper extends VoiceDataDumper
featureDefinition, featureFileReader, halfphoneLRFeatureIndex, numSamples, phoneFeatureIndex, unitDB| Constructor and Description |
|---|
HnmVoiceDataDumper() |
| Modifier and Type | Method and Description |
|---|---|
protected byte[] |
getSamples(Datagram[] datagrams)
Get raw samples from all Datagrams in an array
|
protected HnmTimelineReader |
loadAudioTimeline(String fileName)
Load audio timeline from file
|
static void |
main(String[] args) |
byteswap, byteswap, dumpAudio, dumpData, dumpTextGrid, getDuration, loadFeatureFile, loadUnitDatabaseprotected HnmTimelineReader loadAudioTimeline(String fileName) throws IOException, MaryConfigurationException
Also set the audioFormat needed in getSamples(Datagram[])
loadAudioTimeline in class VoiceDataDumperfileName - to loadIOException - IOExceptionMaryConfigurationException - MaryConfigurationExceptionprotected byte[] getSamples(Datagram[] datagrams) throws IOException
For HnmDatagrams, the samples must be resynthesized from the HntmSpeechFrame in each HnmDatagram. This requires
quite a bit of processing.
getSamples in class VoiceDataDumperdatagrams - whose samples to getIOException - IOExceptionCopyright © 2000–2016 DFKI GmbH. All rights reserved.