Nyaaori/rockbox
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

Remove multithreading support from voicefile creation.

Browse source 
Running TTS and encoders with multiple threads is causing problems on Windows
since introduction of the feature (FS#12106, FS#11994). The current
implementation also makes wrong assumptions (having multiple threads talk to
the SAPI script doesn't make it run faster since it's still one thread
responsible for creation).

Completely remove multithreading support for that for now -- a different
implementation is necessary.

Change-Id: Icafa223644efc370a09186ce28ac83c22902e0c0
This commit is contained in:
Dominik Riebeling 2012-01-14 16:17:13 +01:00
parent c2f0ba7ecd
commit 820dcfdfed
2 changed files with 99 additions and 189 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

257
rbutil/rbutilqt/base/talkgenerator.cpp
View file

@ -21,16 +21,16 @@
#include "systeminfo.h"
#include "wavtrim.h"
TalkGenerator::TalkGenerator(QObject* parent): QObject(parent), encFutureWatcher(this), ttsFutureWatcher(this)
TalkGenerator::TalkGenerator(QObject* parent): QObject(parent)
{
m_userAborted = false;
m_lang = "";
}
//! \brief Creates Talkfiles.
//!
TalkGenerator::Status TalkGenerator::process(QList<TalkEntry>* list,int wavtrimth)
{
m_abort = false;
QString errStr;
bool warnings = false;
@ -103,122 +103,81 @@ TalkGenerator::Status TalkGenerator::process(QList<TalkEntry>* list,int wavtrimt
//!
TalkGenerator::Status TalkGenerator::voiceList(QList<TalkEntry>* list,int wavtrimth)
{
emit logProgress(0, list->size());
int progressMax = list->size();
int m_progress = 0;
emit logProgress(m_progress,progressMax);
QStringList errors;
QStringList duplicates;
m_ttsWarnings = false;
bool warnings = false;
for(int i=0; i < list->size(); i++)
{
(*list)[i].refs.tts = m_tts;
(*list)[i].refs.wavtrim = wavtrimth;
(*list)[i].refs.generator = this;
// enable voice corrections only if a language is set.
if(!m_lang.isEmpty()) {
QString s = (*list)[i].toSpeak;
(*list)[i].toSpeak = correctString(s);
if(m_abort)
{
emit logItem(tr("Voicing aborted"), LOGERROR);
return eERROR;
}
// skip duplicated wav entries
// skip duplicated wav entrys
if(!duplicates.contains(list->at(i).wavfilename))
duplicates.append(list->at(i).wavfilename);
else
{
qDebug() << "[TalkGen] duplicate skipped";
qDebug() << "[TalkGenerator] duplicate skipped";
(*list)[i].voiced = true;
emit logProgress(++m_progress,progressMax);
continue;
}
}
/* If the engine can't be parallelized, we use only 1 thread */
// NOTE: setting the number of maximum threads to use to 1 doesn't seem to
// work as expected -- it causes sporadically output files missing (see
// FS#11994). As a stop-gap solution use a separate implementation in that
// case for running the TTS.
if((m_tts->capabilities() & TTSBase::RunInParallel) != 0)
{
int maxThreadCount = QThreadPool::globalInstance()->maxThreadCount();
qDebug() << "[TalkGenerator] Maximum number of threads used:"
<< QThreadPool::globalInstance()->maxThreadCount();
connect(&ttsFutureWatcher, SIGNAL(progressValueChanged(int)),
this, SLOT(ttsProgress(int)));
ttsFutureWatcher.setFuture(QtConcurrent::map(*list, &TalkGenerator::ttsEntryPoint));
/* We use this loop as an equivalent to ttsFutureWatcher.waitForFinished()
* since the latter blocks all events */
while(ttsFutureWatcher.isRunning())
QCoreApplication::processEvents();
/* Restore global settings, if we changed them */
if ((m_tts->capabilities() & TTSBase::RunInParallel) == 0)
QThreadPool::globalInstance()->setMaxThreadCount(maxThreadCount);
if(ttsFutureWatcher.isCanceled())
return eERROR;
else if(m_ttsWarnings)
return eWARNING;
else
return eOK;
}
else {
qDebug() << "[TalkGenerator] Using single thread TTS workaround";
int items = list->size();
for(int i = 0; i < items; i++) {
if(m_userAborted) {
emit logItem(tr("Voicing aborted"), LOGERROR);
return eERROR;
}
TalkEntry entry = list->at(i);
TalkGenerator::ttsEntryPoint(entry);
(*list)[i] = entry;
emit logProgress(i, items);
}
return m_ttsWarnings ? eWARNING : eOK;
}
}
void TalkGenerator::ttsEntryPoint(TalkEntry& entry)
{
if (!entry.voiced && !entry.toSpeak.isEmpty())
{
QString error;
qDebug() << "[TalkGen] voicing: " << entry.toSpeak << "to" << entry.wavfilename;
TTSStatus status = entry.refs.tts->voice(entry.toSpeak,entry.wavfilename, &error);
if (status == Warning || status == FatalError)
// skip already voiced entrys
if(list->at(i).voiced == true)
{
entry.refs.generator->ttsFailEntry(entry, status, error);
return;
emit logProgress(++m_progress,progressMax);
continue;
}
if (entry.refs.wavtrim != -1)
// skip entry whith empty text
if(list->at(i).toSpeak == "")
{
emit logProgress(++m_progress,progressMax);
continue;
}
// voice entry
QString error;
qDebug() << "[TalkGenerator] voicing: " << list->at(i).toSpeak << "to" << list->at(i).wavfilename;
TTSStatus status = m_tts->voice(list->at(i).toSpeak,list->at(i).wavfilename, &error);
if(status == Warning)
{
warnings = true;
emit logItem(tr("Voicing of %1 failed: %2").arg(list->at(i).toSpeak).arg(error),
LOGWARNING);
}
else if (status == FatalError)
{
emit logItem(tr("Voicing of %1 failed: %2").arg(list->at(i).toSpeak).arg(error),
LOGERROR);
return eERROR;
}
else
(*list)[i].voiced = true;
//wavetrim if needed
if(wavtrimth != -1)
{
char buffer[255];
wavtrim(entry.wavfilename.toLocal8Bit().data(), entry.refs.wavtrim, buffer, 255);
wavtrim(list->at(i).wavfilename.toLocal8Bit().data(),wavtrimth,buffer,255);
}
entry.voiced = true;
emit logProgress(++m_progress,progressMax);
QCoreApplication::processEvents();
}
if(warnings)
return eWARNING;
else
return eOK;
}
void TalkGenerator::ttsFailEntry(const TalkEntry& entry, TTSStatus status, QString error)
{
if(status == Warning)
{
m_ttsWarnings = true;
emit logItem(tr("Voicing of %1 failed: %2").arg(entry.toSpeak).arg(error),
LOGWARNING);
}
else if (status == FatalError)
{
emit logItem(tr("Voicing of %1 failed: %2").arg(entry.toSpeak).arg(error),
LOGERROR);
abort();
}
}
void TalkGenerator::ttsProgress(int value)
{
emit logProgress(value,ttsFutureWatcher.progressMaximum());
}
//! \brief Encodes a List of strings
//!
@ -226,86 +185,58 @@ TalkGenerator::Status TalkGenerator::encodeList(QList<TalkEntry>* list)
{
QStringList duplicates;
int itemsCount = list->size();
emit logProgress(0, itemsCount);
int progressMax = list->size();
int m_progress = 0;
emit logProgress(m_progress,progressMax);
/* Do some preprocessing and remove entries that have not been voiced. */
for (int idx=0; idx < itemsCount; idx++)
for(int i=0; i < list->size(); i++)
{
if(list->at(idx).voiced == false)
if(m_abort)
{
qDebug() << "[TalkGen] unvoiced entry" << list->at(idx).toSpeak <<"detected";
list->removeAt(idx);
itemsCount--;
idx--;
emit logItem(tr("Encoding aborted"), LOGERROR);
return eERROR;
}
//skip non-voiced entrys
if(list->at(i).voiced == false)
{
qDebug() << "non voiced entry" << list->at(i).toSpeak <<"detected";
emit logProgress(++m_progress,progressMax);
continue;
}
if(duplicates.contains(list->at(idx).talkfilename))
//skip duplicates
if(!duplicates.contains(list->at(i).talkfilename))
duplicates.append(list->at(i).talkfilename);
else
{
(*list)[idx].encoded = true; /* make sure we skip this entry */
qDebug() << "[TalkGenerator] duplicate skipped";
(*list)[i].encoded = true;
emit logProgress(++m_progress,progressMax);
continue;
}
duplicates.append(list->at(idx).talkfilename);
(*list)[idx].refs.encoder = m_enc;
(*list)[idx].refs.generator = this; /* not really needed, unless we end up
voicing and encoding with two different
TalkGenerators.*/
//encode entry
qDebug() << "[TalkGenerator] encoding " << list->at(i).wavfilename
<< "to" << list->at(i).talkfilename;
if(!m_enc->encode(list->at(i).wavfilename,list->at(i).talkfilename))
{
emit logItem(tr("Encoding of %1 failed").arg(
QFileInfo(list->at(i).wavfilename).baseName()), LOGERROR);
return eERROR;
}
(*list)[i].encoded = true;
emit logProgress(++m_progress,progressMax);
QCoreApplication::processEvents();
}
connect(&encFutureWatcher, SIGNAL(progressValueChanged(int)),
this, SLOT(encProgress(int)));
encFutureWatcher.setFuture(QtConcurrent::map(*list, &TalkGenerator::encEntryPoint));
/* We use this loop as an equivalent to encFutureWatcher.waitForFinished()
* since the latter blocks all events */
while (encFutureWatcher.isRunning())
QCoreApplication::processEvents(QEventLoop::AllEvents);
if (encFutureWatcher.isCanceled())
return eERROR;
else
return eOK;
return eOK;
}
void TalkGenerator::encEntryPoint(TalkEntry& entry)
{
if(!entry.encoded)
{
bool res = entry.refs.encoder->encode(entry.wavfilename, entry.talkfilename);
entry.encoded = res;
if (!entry.encoded)
entry.refs.generator->encFailEntry(entry);
}
return;
}
void TalkGenerator::encProgress(int value)
{
emit logProgress(value, encFutureWatcher.progressMaximum());
}
void TalkGenerator::encFailEntry(const TalkEntry& entry)
{
emit logItem(tr("Encoding of %1 failed").arg(
QFileInfo(entry.wavfilename).baseName()), LOGERROR);
abort();
}
//! \brief slot, which is connected to the abort of the Logger. Sets a flag, so Creating Talkfiles ends at the next possible position
//! \brief slot, which is connected to the abort of the Logger.
//Sets a flag, so Creating Talkfiles ends at the next possible position
//!
void TalkGenerator::abort()
{
if (ttsFutureWatcher.isRunning())
{
ttsFutureWatcher.cancel();
emit logItem(tr("Voicing aborted"), LOGERROR);
}
if (encFutureWatcher.isRunning())
{
encFutureWatcher.cancel();
emit logItem(tr("Encoding aborted"), LOGERROR);
}
m_userAborted = true;
m_abort = true;
}
QString TalkGenerator::correctString(QString s)
@ -325,9 +256,9 @@ QString TalkGenerator::correctString(QString s)
qDebug() << "[VoiceFileCreator] corrected string" << s << "to" << corrected;
return corrected;
m_abort = true;
}
void TalkGenerator::setLang(QString name)
{
m_lang = name;

31
rbutil/rbutilqt/base/talkgenerator.h
View file

@ -48,30 +48,15 @@ public:
QString target;
bool voiced;
bool encoded;
/* We need the following members because
* 1) the QtConcurrent entry points are all static methods (and we
* need to communicate with the TalkGenerator)
* 2) we are not guaranteed to go through the list in any
* particular order, so we can't use the progress slot
* for error checking */
struct
{
EncoderBase* encoder;
TTSBase* tts;
TalkGenerator* generator;
int wavtrim;
} refs;
};
TalkGenerator(QObject* parent);
Status process(QList<TalkEntry>* list,int wavtrimth = -1);
QString correctString(QString s);
public slots:
void abort();
void encProgress(int value);
void ttsProgress(int value);
void setLang(QString name);
signals:
@ -80,22 +65,12 @@ signals:
void logProgress(int, int); //! set progress bar.
private:
QFutureWatcher<void> encFutureWatcher;
QFutureWatcher<void> ttsFutureWatcher;
void encFailEntry(const TalkEntry& entry);
void ttsFailEntry(const TalkEntry& entry, TTSStatus status, QString error);
Status voiceList(QList<TalkEntry>* list,int wavetrimth);
Status encodeList(QList<TalkEntry>* list);
static void encEntryPoint(TalkEntry& entry);
static void ttsEntryPoint(TalkEntry& entry);
TTSBase* m_tts;
EncoderBase* m_enc;
bool m_ttsWarnings;
bool m_userAborted;
QString m_lang;
struct CorrectionItems
@ -105,6 +80,10 @@ private:
QString modifier;
};
QList<struct CorrectionItems> m_corrections;
bool m_abort;
};