def __write_trs_track(track_filename, track_content, duration):
begin = TimePoint(0.)
end = TimePoint(duration)
ann = Annotation(TimeInterval(begin, end), Label(track_content))
trs = Transcription()
tier = trs.NewTier("Transcription")
tier.Append(ann)
sppas.src.annotationdata.aio.write(track_filename, trs)
def __init__(self, name="NoName", mintime=0., maxtime=0.):
"""
Creates a new SegmentsIn instance.
"""
Transcription.__init__(self, name, mintime, maxtime)
self.alignerio = AlignerIO()
self._radius = 0.005
self._tracknames = TrackNamesGenerator()
def __init__(self, name="NoName", mintime=0., maxtime=0.):
"""
Creates a new SegmentsIn instance.
"""
Transcription.__init__(self, name, mintime, maxtime)
self.alignerio = AlignerIO()
self._radius = 0.005
self._tracknames = TrackNamesGenerator()
def __init__(self, name="NoName", mintime=0., maxtime=0.):
""" Creates a new TrackSplitter instance.
:param name: (str)
"""
Transcription.__init__(self, name, mintime, maxtime)
self._radius = 0.005
self._tracknames = TrackNamesGenerator()
self._aligntrack = None
def __init__(self, name="NoName", mintime=0., maxtime=0.):
""" Creates a new TrackSplitter instance.
:param name: (str)
"""
Transcription.__init__(self, name, mintime, maxtime)
self._radius = 0.005
self._tracknames = TrackNamesGenerator()
self._aligntrack = None
def __init__(self, name="AnnotationSystemDataSet"):
"""
Initialize a new instance.
@type name: str
@param name: the name of the transcription
"""
Transcription.__init__(self, name)
self.__id_tier_map = {}
def set_transcription(self, trs):
""" Set a new Transcription.
:param trs: (Transcription) Input transcription from which it's
possible to extract IPUs.
"""
if trs is not None:
self._trsinput = trs
else:
self._trsinput = Transcription()
def run(self, input_filename1, input_filename2, output_filename):
""" Run the Repetition Automatic Detection annotation.
:param input_filename1: Name of the file with aligned tokens of spkeaker 1 (the source)
:param input_filename2: Name of the file with aligned tokens of spkeaker 2 (the echo) if OR, or None for SR
:param output_filename: Name of the file to save the result
"""
self.print_filename(input_filename1)
self.print_options()
self.print_diagnosis(input_filename1)
if input_filename2 is not None:
self.print_diagnosis(input_filename2)
if self.logfile is not None:
self.logfile.print_message("Span = " + str(self._span), indent=3)
self.logfile.print_message("Alpha = " + str(self._alpha), indent=3)
# Get the tiers to be used
# ---------------------------------------------------------------
# Tokens of main speaker
trs_input1 = sppas.src.annotationdata.aio.read(input_filename1)
tier1 = sppasSearchTier.aligned_tokens(trs_input1)
if tier1.IsEmpty() is True:
raise Exception("Empty tokens tier (main speaker).\n")
# Tokens of echoing speaker (if any)
tier2 = None
if input_filename2 is not None:
trs_input2 = sppas.src.annotationdata.aio.read(input_filename2)
tier2 = sppasSearchTier.aligned_tokens(trs_input2)
if tier2.IsEmpty() is True:
raise Exception("Empty tokens tier (echoing speaker).\n")
# Lemmatize input?
if self._use_lemmatize:
tier1 = self.lemmatize(tier1)
if tier2 is not None:
tier2 = self.lemmatize(tier2)
# Repetition Automatic Detection
# ---------------------------------------------------------------
if tier2 is None:
(src_tier, echo_tier) = self.self_detection(tier1)
else:
(src_tier, echo_tier) = self.other_detection(tier1, tier2)
# Save results
# --------------------------------------------------------------
trs_output = Transcription("Repetitions")
trs_output.Append(src_tier)
trs_output.Append(echo_tier)
sppas.src.annotationdata.aio.write(output_filename, trs_output)
def __init__(self, trs):
""" Creates a new IPUsTrs instance.
:param trs: (Transcription) Input transcription from which it's
possible to extract IPUs.
Expected tiers are:
- first tier: the IPUs content [required]
- second tier: the IPUs file names [optional]
"""
super(IPUsTrs, self).__init__()
self._trsinput = Transcription()
self._units = list() # List of the content of the units (if any)
self._names = list() # List of file names for IPUs (if any)
self.set_transcription(trs)
def print_targets(self, targets, output_filename=None, trs=None):
""" Print the set of selected targets.
:param targets:
:param output_filename: (str)
:param trs: (Transcription)
"""
if output_filename is not None:
if output_filename is "STDOUT":
output = sys.stdout
self.__print_tgts(targets, output)
elif output_filename.lower().endswith('momel') is True:
output = open(output_filename, "w")
self.__print_tgts(targets, output)
output.close()
if trs is not None:
# Attention: time in targets is in milliseconds!
tier = trs.NewTier(name="Momel")
for i in range(len(targets)):
_time = targets[i].get_x() * (0.001 * self.PAS_TRAME)
_label = str("%d" % (targets[i].get_y()))
try:
tier.Append(Annotation(TimePoint(_time), Label(_label)))
except Exception:
if self.logfile is not None:
self.logfile.print_message("Ignore target: time=" +
str(_time) + " and value=" +
_label,
indent=2,
status=3)
if output_filename is not None and output_filename.lower(
).endswith('.pitchtier'):
trsp = Transcription()
trsp.Add(tier)
try:
sppas.src.annotationdata.aio.write(output_filename, trsp)
except Exception:
if self.logfile is not None:
self.logfile.print_message(
"Can't write PitchTier output file.", status=-1)
return tier
def LoadFile(self, filename):
"""
Load a file in memory and show it.
@param filename is an annotated file.
"""
self._filename = filename
if os.path.exists(filename) is False:
self._transcription = Transcription("Empty")
return
try:
self._transcription = sppas.src.annotationdata.aio.read(filename)
self._dirty = False
self._boxtitle.SetForegroundColour(FG_FILE_COLOUR)
self.Refresh()
except Exception as e:
logging.info('Error loading file %s: %s' % (filename, str(e)))
self._transcription = Transcription("IO-Error")
class TrsList(wx.Panel):
"""
:author: Brigitte Bigi
:organization: Laboratoire Parole et Langage, Aix-en-Provence, France
:contact: [email protected]
:license: GPL, v3
:copyright: Copyright (C) 2011-2018 Brigitte Bigi
:summary: Show data about transcriptions, in a panel including a list of tiers.
"""
def __init__(self, parent, filename, trs=None, multiple=False):
wx.Panel.__init__(self, parent, -1, size=wx.DefaultSize)
# initialize the GUI
self._prefs = Preferences()
self._filename = filename
self._dirty = False # the transcription was changed
self._selected = False # the transcription is selected
self._protected = [
] # list of the tiers that are protected (can't be modified)
if len(filename) == 0:
self._filename = "Empty"
boxtitle = self._create_title()
self.tier_list = self._create_list(multiple)
# load the Transcription
if trs is None and len(filename) != 0:
self.LoadFile(filename)
else:
self._transcription = trs
# add Transcription information in the list
for i in range(self._transcription.GetSize()):
self.SetTierProperties(i)
self._checksize()
# events
self.Bind(wx.EVT_LIST_ITEM_SELECTED, self.OnListItemSelected,
self.tier_list)
self.Bind(wx.EVT_LIST_COL_CLICK, self.OnListItemSelected,
self.tier_list)
# layout
sizer = wx.BoxSizer(wx.VERTICAL)
sizer.Add(boxtitle, 0, wx.EXPAND | wx.ALL, border=4)
sizer.Add(self.tier_list, 1, wx.EXPAND | wx.ALL, border=4)
self.SetFont(self._prefs.GetValue('M_FONT'))
self.SetForegroundColour(self._prefs.GetValue('M_FG_COLOUR'))
self.SetBackgroundColour(self._prefs.GetValue('M_BG_COLOUR'))
self._boxtitle.SetForegroundColour(FG_FILE_COLOUR)
self.SetSizerAndFit(sizer)
self.SetAutoLayout(True)
self.Layout()
# ----------------------------------------------------------------------
def _create_title(self):
""" Create the title of the panel. """
_sizer = wx.BoxSizer(wx.HORIZONTAL)
self._static_tx = wx.TextCtrl(self,
-1,
"File: ",
style=wx.TE_READONLY | wx.NO_BORDER)
self._boxtitle = wx.TextCtrl(self,
-1,
self._filename,
style=wx.TE_READONLY | wx.NO_BORDER)
_sizer.Add(self._static_tx, 0, wx.RIGHT, border=2)
_sizer.Add(self._boxtitle, 1, wx.EXPAND)
return _sizer
# ----------------------------------------------------------------------
def _create_list(self, multiple=False):
""" Create the list to show information of a each tier of a transcription. """
if multiple:
tier_list = CheckListCtrl(self,
-1,
style=wx.LC_REPORT | wx.BORDER_NONE)
else:
tier_list = CheckListCtrl(self,
-1,
style=wx.LC_REPORT | wx.BORDER_NONE
| wx.LC_SINGLE_SEL)
# Add all columns
col_names = [
" Nb ", " Name ", " Begin ", " End ", " Type ",
" Size "
]
for i, n in enumerate(col_names):
tier_list.InsertColumn(i, n)
# Fix column width
for i in range(len(col_names)):
tier_list.SetColumnWidth(i, wx.LIST_AUTOSIZE_USEHEADER)
# Enlarge column with tier name
tier_list.SetColumnWidth(1, 140)
return tier_list
# -------------------------------------------------------------------------
def SetTierProperties(self, tier_idx):
""" Display tier properties. """
try:
tier = self._transcription[tier_idx]
if tier.IsPoint() is True:
tier_type = "Point"
elif tier.IsInterval():
tier_type = "Interval"
elif tier.IsDisjoint():
tier_type = "Disjoint"
else:
tier_type = "Unknown"
if tier.IsEmpty() is True:
begin = " ... "
end = " ... "
else:
begin = str(tier.GetBeginValue())
end = str(tier.GetEndValue())
self.tier_list.InsertStringItem(tier_idx,
"Tier %d" % (tier_idx + 1))
self.tier_list.SetStringItem(tier_idx, 1, tier.GetName())
self.tier_list.SetStringItem(tier_idx, 2, begin)
self.tier_list.SetStringItem(tier_idx, 3, end)
self.tier_list.SetStringItem(tier_idx, 4, tier_type)
self.tier_list.SetStringItem(tier_idx, 5, str(tier.GetSize()))
except Exception as e:
self.tier_list.InsertStringItem(1, "Error: " + str(e))
# ----------------------------------------------------------------------
# Callbacks...
# ----------------------------------------------------------------------
def OnListItemSelected(self, event):
""" An item of this panel was clicked. Inform the parent. """
evt = PanelSelectedEvent(panel=self)
evt.SetEventObject(self)
wx.PostEvent(self.GetParent(), evt)
# ----------------------------------------------------------------------
# GUI
# ----------------------------------------------------------------------
def SetPreferences(self, prefs):
""" Set new preferences. """
self._prefs = prefs
self.SetBackgroundColour(self._prefs.GetValue("M_BG_COLOUR"))
self.SetForegroundColour(self._prefs.GetValue("M_FG_COLOUR"))
self.SetFont(self._prefs.GetValue("M_FONT"))
# -------------------------------------------------------------------------
def SetFont(self, font):
""" Set a new font. """
wx.Window.SetFont(self, font)
self.tier_list.SetFont(font)
for i in range(self._transcription.GetSize()):
self.tier_list.SetItemFont(i, font)
self._static_tx.SetFont(font)
self._boxtitle.SetFont(font)
self.Layout() # bigger/smaller font can impact on the layout
# -------------------------------------------------------------------------
def SetBackgroundColour(self, color):
""" Set background. """
wx.Window.SetBackgroundColour(self, color)
self.tier_list.SetBackgroundColour(color)
for i in range(self._transcription.GetSize()):
self.tier_list.SetItemBackgroundColour(i, color)
self._static_tx.SetBackgroundColour(color)
self._boxtitle.SetBackgroundColour(color)
self.Refresh()
# -------------------------------------------------------------------------
def SetForegroundColour(self, color):
""" Set foreground and items text color. """
wx.Window.SetForegroundColour(self, color)
self.tier_list.SetForegroundColour(color)
for i in range(self._transcription.GetSize()):
self.tier_list.SetItemTextColour(i, color)
self._static_tx.SetForegroundColour(color)
self.Refresh()
# ----------------------------------------------------------------------
# Functions...
# ----------------------------------------------------------------------
def Protect(self):
"""
Fix the current list of tiers as protected: they won't be changed.
"""
self._protected = []
for i, t in enumerate(self._transcription):
self._protected.append(t)
self.tier_list.SetItemTextColour(i, wx.Colour(140, 10, 10))
# -------------------------------------------------------------------------
def Unprotect(self):
"""
Erase the list of protected tiers.
"""
self._protected = []
# ----------------------------------------------------------------------
def IsSelected(self, tiername, case_sensitive=False):
"""
Return True if the tier is selected.
"""
i = self._transcription.GetIndex(tiername, case_sensitive)
if i != -1:
return self.tier_list.IsSelected(i)
return False
# ----------------------------------------------------------------------
def Select(self, tiername, case_sensitive=False):
"""
Select tiers which name is exactly matching.
"""
i = self._transcription.GetIndex(tiername, case_sensitive)
if i != -1:
self.tier_list.Select(i, on=True)
return True
return False
# ----------------------------------------------------------------------
def Deselect(self):
#for i in range(self.tier_list.GetItemCount()):
# self.tier_list.Select(i, on=0)
self.tier_list.DeSelectAll()
# ----------------------------------------------------------------------
def Rename(self):
""" Rename the selected tier. Dialog with the user to get the new name. """
if self._transcription.GetSize() == 0:
return
# Get the selected tier in the list
sellist = self.tier_list.GetFirstSelected()
# Nothing selected
if sellist == -1:
return
# Too many selected items
if self.tier_list.GetSelectedItemCount() > 1:
ShowInformation(self,
self._prefs,
'Only one tier has to be checked to be renamed...',
style=wx.ICON_INFORMATION)
return
tier = self._transcription[sellist]
if tier in self._protected:
ShowInformation(self,
self._prefs,
"Attempt to rename a protected tier: forbidden!",
style=wx.ICON_INFORMATION)
return
# Ask the user to enter a new name
dlg = wx.TextEntryDialog(self, 'Indicate the new tier name',
'Data Roamer', 'Rename a tier.')
dlg.SetValue(self._transcription[sellist].GetName())
if dlg.ShowModal() == wx.ID_OK:
# Update tier name of the transcription
tier.SetName(dlg.GetValue())
# Update tier name of the list
self.tier_list.SetStringItem(sellist, 1, dlg.GetValue())
self._dirty = True
self._boxtitle.SetForegroundColour(FG_FILE_DIRTY_COLOUR)
self.Refresh()
dlg.Destroy()
# ----------------------------------------------------------------------
def Cut(self):
""" Cut the selected tier. Return the clipboard. """
if self._transcription.GetSize() == 0:
return
# Get the selected tier in the list
sellist = self.tier_list.GetFirstSelected()
# No tier selected
if sellist == -1:
return
# Too many selected items
if self.tier_list.GetSelectedItemCount() > 1:
ShowInformation(self,
self._prefs,
'One tier must be checked.',
style=wx.ICON_INFORMATION)
return
# Copy the tier to the clipboard
tier = self._transcription[sellist]
if tier in self._protected:
ShowInformation(self,
self._prefs,
"Attempt to cut a protected tier: forbidden!",
style=wx.ICON_INFORMATION)
return
clipboard = tier.Copy()
# Delete tier of the transcription
self._transcription.Remove(sellist)
# Delete tier of the list
self.tier_list.DeleteItem(sellist)
# Update tier numbers of next items in the list.
for i in range(sellist, self.tier_list.GetItemCount()):
self.tier_list.SetStringItem(i, 0, "Tier " + str(i + 1))
self.Deselect()
self._checksize()
self._dirty = True
self._boxtitle.SetForegroundColour(FG_FILE_DIRTY_COLOUR)
self.Refresh()
return clipboard
# ----------------------------------------------------------------------
def Copy(self):
""" Return the selected tier. """
if self._transcription.GetSize() == 0:
return
# Get the selected tier in the list
sellist = self.tier_list.GetFirstSelected()
if sellist == -1:
return
# Too many selected items
if self.tier_list.GetSelectedItemCount() > 1:
ShowInformation(self,
self._prefs,
"One tier must be checked",
style=wx.ICON_INFORMATION)
return
# Copy the tier to the clipboard
tier = self._transcription[sellist]
return tier.Copy()
# ----------------------------------------------------------------------
def Paste(self, clipboard):
""" Paste the clipboard tier to the current page. """
# Get the clipboard tier
if clipboard is None:
return
# Append clipboard to the transcription
tier = clipboard #.Copy()
self.Append(tier)
# The tier comes from another Transcription... must update infos.
if not (tier.GetTranscription() is self._transcription):
# parent transcription
tier.SetTranscription(self._transcription)
# And if CtrlVocab...
# TODO
self._checksize()
# ----------------------------------------------------------------------
def Delete(self):
""" Delete the selected tier.
Dialog with the user to confirm. """
if self._transcription.GetSize() == 0:
return 0
# Get the selected tier in the list of this page
sellist = self.tier_list.GetFirstSelected()
if sellist == -1:
return 0
# Get Indexes of tiers to remove
indexes = []
while sellist != -1:
indexes.append(sellist)
sellist = self.tier_list.GetNextSelected(sellist)
# Ask the user to confirm before deleting
delete = 0
message = 'Are you sure you want to definitively delete:\n' \
'%d tiers in %s?' % (len(indexes), self._filename)
dlg = ShowYesNoQuestion(self, self._prefs, message)
if dlg == wx.ID_YES:
for sellist in reversed(sorted(indexes)):
item = self.tier_list.GetItem(sellist)
tier = self._transcription[sellist]
if tier in self._protected:
pass
else:
# Delete tier of the transcription
self._transcription.Remove(sellist)
# Delete tier of the list
self.tier_list.DeleteItem(sellist)
delete = delete + 1
# Update tier numbers of next items in the list.
for i in range(sellist, self.tier_list.GetItemCount()):
self.tier_list.SetStringItem(i, 0, str(i + 1))
self._dirty = True
self._boxtitle.SetForegroundColour(FG_FILE_DIRTY_COLOUR)
self.Refresh
self._checksize()
return delete
# ----------------------------------------------------------------------
def Duplicate(self):
""" Duplicate the selected tier. """
if self._transcription.GetSize() == 0:
return
# Get the selected tier index in the list
sellist = self.tier_list.GetFirstSelected()
if sellist == -1:
return
# Too many selected items
if self.tier_list.GetSelectedItemCount() > 1:
ShowInformation(self,
self._prefs,
"One tier must be checked",
style=wx.ICON_INFORMATION)
return
tier = self._transcription[sellist]
self.Append(tier.Copy())
# ----------------------------------------------------------------------
def MoveUp(self):
""" Move up the selected tier (except for the first one). """
if self._transcription.GetSize() == 0:
return
# Get the selected tier in the list
sellist = self.tier_list.GetFirstSelected()
if sellist == -1:
return
# Too many selected items
if self.tier_list.GetSelectedItemCount() > 1:
ShowInformation(self,
self._prefs,
"One tier must be checked",
style=wx.ICON_INFORMATION)
return
#
tier = self._transcription[sellist]
if tier in self._protected:
ShowInformation(self,
self._prefs,
"Attempt to move a protected tier: forbidden!",
style=wx.ICON_INFORMATION)
return
#Impossible to move up the first tier.
if sellist == 0:
return
# Pop selected tier from transcription.
try:
self._transcription._hierarchy.remove_tier(
self._transcription[sellist]
) # waiting a better way to work with hierarchy...
except Exception:
pass
self._transcription.Pop(sellist)
# Delete old tier of the list
self.tier_list.DeleteItem(sellist)
# Add tier to the transcription
tierindex = self._transcription.Add(tier, sellist - 1)
# Add tier to the list
self.SetTierProperties(tierindex)
# Update tier number
self.tier_list.SetStringItem(sellist, 0, str(sellist + 1))
# Let the item selected
self.tier_list.Select(sellist - 1, on=True)
self._dirty = True
self._boxtitle.SetForegroundColour(FG_FILE_DIRTY_COLOUR)
self.Refresh()
# ----------------------------------------------------------------------
def MoveDown(self):
""" Move down the selected tier (except for the last one). """
if self._transcription.GetSize() == 0:
return
# Get the selected tier in the list
sellist = self.tier_list.GetFirstSelected()
if sellist == -1:
return
# Too many selected items
if self.tier_list.GetSelectedItemCount() > 1:
ShowInformation(self,
self._prefs,
"One tier must be checked",
style=wx.ICON_INFORMATION)
return
#
tier = self._transcription[sellist]
if tier in self._protected:
ShowInformation(self,
self._prefs,
"Attempting to move a protected tier: forbidden!",
style=wx.ICON_INFORMATION)
return
# Impossible to move down the last tier.
if (sellist + 1) == self.tier_list.GetItemCount():
return
# Pop selected tier from transcription.
try:
self._transcription._hierarchy.remove_tier(
self._transcription[sellist]
) # waiting a better way to work with hierarchy...
except Exception:
pass
self._transcription.Pop(sellist)
# Delete old tier of the list
self.tier_list.DeleteItem(sellist)
# Add tier to the transcription
if (sellist + 1) >= self.tier_list.GetItemCount():
tierindex = self._transcription.Add(tier)
else:
tierindex = self._transcription.Add(tier, sellist + 1)
# Add tier to the list
self.SetTierProperties(tierindex)
# Update tier number
self.tier_list.SetStringItem(sellist, 0, "Tier " + str(sellist + 1))
self.tier_list.SetStringItem(sellist + 1, 0,
"Tier " + str(tierindex + 1))
# Let the item selected
self.tier_list.Select(sellist + 1, on=True)
self._dirty = True
self._boxtitle.SetForegroundColour(FG_FILE_DIRTY_COLOUR)
self.Refresh()
# ----------------------------------------------------------------------
def Radius(self):
""" Fix a new radius value to all TimePoint instances of the selected tier. """
if self._transcription.GetSize() == 0:
return
# Get the selected tier in the list
sellist = self.tier_list.GetFirstSelected()
if sellist == -1:
return
#
tier = self._transcription[sellist]
if tier in self._protected:
ShowInformation(self,
self._prefs,
"Attempt to modify a protected tier: forbidden!",
style=wx.ICON_INFORMATION)
return
# Open a dialog to ask the new radius value
radius = tier.GetBegin().GetRadius()
dlg = RadiusChooser(self, self._prefs, radius)
if dlg.ShowModal() == wx.ID_OK:
# Get the value
r = dlg.GetValue()
try:
r = float(r)
if r > 1.0:
raise ValueError('Radius must range 0-1.')
except:
logging.info('Radius cancelled (can not be applied: %f).' % r)
return
# Set the value
while sellist != -1:
tier.SetRadius(r)
logging.debug('Radius fixed to %f' % r)
sellist = self.tier_list.GetNextSelected(sellist)
dlg.Destroy()
# ----------------------------------------------------------------------
def Preview(self):
""" Open a grid frame with the selected tier content. """
if self._transcription.GetSize() == 0:
return
# Get the selected tier in the list
sellist = self.tier_list.GetFirstSelected()
if sellist == -1:
return
# Too many selected items
if self.tier_list.GetSelectedItemCount() > 1:
ShowInformation(self,
self._prefs,
"One tier only must be checked",
style=wx.ICON_INFORMATION)
return
tier = self._transcription[sellist]
dlg = PreviewTierDialog(self, self._prefs, tiers=[tier])
dlg.Show()
# ----------------------------------------------------------------------
def Append(self, newtier):
"""
Append a tier in the transcription and in the list.
"""
# Append tier to the transcription
tierindex = self._transcription.Append(newtier)
# Append tier to the list
self.SetTierProperties(tierindex)
# Display information
self._dirty = True
self._boxtitle.SetForegroundColour(FG_FILE_DIRTY_COLOUR)
self.Refresh()
# ----------------------------------------------------------------------
def LoadFile(self, filename):
"""
Load a file in memory and show it.
@param filename is an annotated file.
"""
self._filename = filename
if os.path.exists(filename) is False:
self._transcription = Transcription("Empty")
return
try:
self._transcription = sppas.src.annotationdata.aio.read(filename)
self._dirty = False
self._boxtitle.SetForegroundColour(FG_FILE_COLOUR)
self.Refresh()
except Exception as e:
logging.info('Error loading file %s: %s' % (filename, str(e)))
self._transcription = Transcription("IO-Error")
#raise
# ----------------------------------------------------------------------
def Save(self):
""" Save the current page content. """
if self._dirty is False:
return
try:
sppas.src.annotationdata.aio.write(self._filename,
self._transcription)
self._dirty = False
self._boxtitle.SetForegroundColour(FG_FILE_COLOUR)
self.Refresh()
except Exception as e:
# give information
ShowInformation(self,
self._prefs,
'File not saved: %s' % str(e),
style=wx.ICON_ERROR)
# ----------------------------------------------------------------------
def SaveAs(self, filename):
"""
Save the current page content with another file name.
Keep everything un-changed in self.
"""
try:
sppas.src.annotationdata.aio.write(filename, self._transcription)
except Exception as e:
# give information
ShowInformation(self,
self._prefs,
'File not saved: %s' % str(e),
style=wx.ICON_ERROR)
# ----------------------------------------------------------------------
def GetTranscription(self):
""" Return the Transcription. """
return self._transcription
# ----------------------------------------------------------------------
def GetTranscriptionName(self):
""" Return the name of the transcription. """
return self._transcription.GetName()
# ----------------------------------------------------------------------
# Private
# ----------------------------------------------------------------------
def _checksize(self):
"""
Check the transcription size. Append an "empty line" if
transcription is empty. Remove this empty line if transcription
is not empty. Return True if something has changed.
"""
# Append an "empty" line in the ListCtrl
if self._transcription.GetSize() == 0 and self.tier_list.GetItemCount(
) == 0:
self.tier_list.InsertStringItem(0, " ... ")
if self._transcription.GetName() == "IO-Error":
self.tier_list.SetStringItem(
0, 1, " Error while reading this file ")
else:
self.tier_list.SetStringItem(0, 1, " Empty file: no tiers ")
for i in range(2, 5):
self.tier_list.SetStringItem(0, i, " ")
return True
# Remove the "empty" line of the ListCtrl
if self._transcription.GetSize() < self.tier_list.GetItemCount():
self.tier_list.DeleteItem(self.tier_list.GetItemCount() - 1)
return True
return False
def tracks2transcription(self, ipustrs, ipusaudio, add_ipu_idx=False):
""" Create a Transcription object from tracks.
:param ipustrs: (IPUsTrs)
:param ipusaudio: (IPUsAudio)
:param add_ipu_idx: (bool)
"""
if len(self.tracks) == 0:
raise IOError('No IPUs to write.\n')
# Extract the info we need from IPUsAudio
framerate = ipusaudio.get_channel().get_framerate()
end_time = ipusaudio.get_channel().get_duration()
# Extract the info we need from ipustrs
try:
medialist = ipustrs.trsinput.GetMedia()
if len(medialist) > 0:
media = medialist[0]
else:
media = None
except Exception:
media = None
units = ipustrs.get_units()
if len(units) != 0:
if len(self.tracks) != len(units):
raise Exception('Inconsistent number of tracks and units. '
'Got %d audio tracks, and %d units.\n' %
(len(self.tracks), len(units)))
# Create the transcription and tiers
trs = Transcription("IPU-Segmentation")
tieripu = trs.NewTier("IPUs")
tier = trs.NewTier("Transcription")
radius = ipusaudio.get_win_length() / 8.
# vagueness is win_length divided by 4 (see "refine" method of sppasChannelSilence class)
# radius is vagueness divided by 2
# Convert the tracks: from frames to times
tracks_times = frames2times(self.tracks, framerate)
i = 0
to_prec = 0.
for (from_time, to_time) in tracks_times:
# From the previous track to the current track: silence
if to_prec < from_time:
begin = to_prec
end = from_time
a = Annotation(
TimeInterval(TimePoint(begin, radius),
TimePoint(end, radius)), Label("#"))
tieripu.Append(a)
tier.Append(a.Copy())
# New track with speech
begin = from_time
end = to_time
# ... IPU tier
label = "ipu_%d" % (i + 1)
a = Annotation(
TimeInterval(TimePoint(begin, radius), TimePoint(end, radius)),
Label(label))
tieripu.Append(a)
# ... Transcription tier
if add_ipu_idx is False:
label = ""
if len(units) > 0:
label = label + " " + units[i]
a = Annotation(
TimeInterval(TimePoint(begin, radius), TimePoint(end, radius)),
Label(label))
tier.Append(a)
# Go to the next
i += 1
to_prec = to_time
# The end is a silence?
if to_prec < end_time:
begin = TimePoint(to_prec, radius)
end = TimePoint(end_time, radius)
if begin < end:
a = Annotation(TimeInterval(begin, end), Label("#"))
tieripu.Append(a)
tier.Append(a.Copy())
# Link both tiers: IPU and Transcription
try:
trs.GetHierarchy().add_link('TimeAssociation', tieripu, tier)
except Exception:
pass
# Set media
if media is not None:
trs.AddMedia(media)
for tier in trs:
tier.SetMedia(media)
return trs
def get_transcription(self, input_filename, tier_idx=None):
""" Extract transcription from a file, either time-aligned or not.
If input is a simple text file, it must be formatted like:
- each line is supposed to be at least one unit;
- each '#' symbol is considered as a unit boundary.
- both can be combined.
If input is a time-aligned file, the expected tier name for the
transcription are:
- priority: trans in the tier name;
- secondary: trs, ortho, toe or ipu in the tier name.
It also extracts IPUs file names if any, i.e. a tier with name
"Name" or "File".
:param input_filename: (str) Name of the input file
:param tier_idx: (int) Force the tier index for the transcription
:returns: Transcription
"""
if input_filename is None:
return Transcription()
trs_input = sppas.src.annotationdata.aio.read(input_filename)
# input is a simple text file
if input_filename.lower().endswith("txt"):
if trs_input.GetSize() != 1:
raise IOError(
'Error while reading file (expected one tier. Got %d)' %
trs_input.GetSize())
return trs_input
# input is a time-aligned file
if tier_idx is None:
trs_tier = sppasSearchTier.transcription(trs_input)
else:
trs_tier = trs_input[tier_idx]
trs_output = Transcription("Output")
if self.logfile:
self.logfile.print_message("IPUs+Transcription tier found: %s" %
trs_tier.GetName(),
indent=3,
status=INFO_ID)
trs_tier.SetName('Transcription')
trs_output.Append(trs_tier)
# Expected IPUs file names
for tier in trs_input:
tier_name = tier.GetName().lower()
if "name" in tier_name or "file" in tier_name:
if self.logfile:
self.logfile.print_message("IPUs file names found: %s" %
tier.GetName(),
indent=3,
status=INFO_ID)
tier.SetName("Name")
trs_output.Append(tier)
break
return trs_output
# ----------------------------------------------------------------------------
# Load input data
mapping = TierMapping(args.m)
mapping.set_reverse(False) # from sampa to ipa direction
mapping.set_keep_miss(True) # keep unknown entries as given
mapping.set_miss_symbol("") # not used!
mapping.set_delimiters([]) # longest matching
# read content
trs_input = aio.read(args.i)
# ----------------------------------------------------------------------------
# Convert input file
trs = Transcription(name=trs_input.GetName()+"-IPA")
for n in args.n.split(','):
print(" -> Tier {:s}:".format(n))
tier = trs_input.Find(n, case_sensitive=False)
if tier is not None:
new_tier = mapping.map_tier(tier)
new_tier.SetName(n+"-IPA")
new_tier.metadata = tier.metadata
trs.Append(new_tier)
else:
print(" [IGNORED] Wrong tier name.")
# Set the other members
trs.metadata = trs_input.metadata
if tieridx < 0 or tieridx > trs.GetSize():
print('Error: Bad tier number.\n')
sys.exit(1)
tier = trs[tieridx]
if not mode:
mode.append(0)
d = {0: 'exact', 1: 'contains', 2: 'startswith', 3: 'endswith', 4: 'regexp'}
prefix = "" if "CASE_SENSITIVE" in options else "i"
bools = [Sel(**{prefix + d[key]: p}) for key in mode for p in patterns]
pred = functools.reduce(operator.or_, bools)
pred = ~pred if "REVERSE" in options else pred
filtered_annotations = filter(pred, tier)
if not filtered_annotations:
print("NO RESULT")
sys.exit(0)
filteredtier = Tier(tier.Name)
for a in filtered_annotations:
filteredtier.Add(a)
if fileoutput is None:
for a in filteredtier:
print(a)
else:
trs = Transcription()
trs.Add(filteredtier)
sppas.src.annotationdata.aio.write(fileoutput, trs)
def __init__(self, name="NoName", mintime=0., maxtime=0.):
""" Creates a new Phonedit Transcription instance. """
Transcription.__init__(self, name, mintime, maxtime)
def run(self, input_filename, trsoutput=None, outputfile=None):
"""
Apply momel from a pitch file.
"""
self.print_filename(input_filename)
self.print_options()
self.print_diagnosis(input_filename)
# Get pitch values from the input
pitch = self.set_pitch(input_filename)
# Selected values (Target points) for this set of pitch values
targets = []
# List of pitch values of one **estimated** Inter-Pausal-Unit (ipu)
ipupitch = []
# Number of consecutive null F0 values
nbzero = 0
# Current time value
curtime = 0
# For each f0 value of the wav file
for p in pitch:
if p == 0:
nbzero += 1
else:
nbzero = 0
ipupitch.append(p)
# If the number of null values exceed 300ms,
# we consider this is a silence and estimate Momel
# on the recorded list of pitch values of the **estimated** IPU.
if (nbzero * self.PAS_TRAME) > 299:
if len(ipupitch) > 0 and (len(ipupitch) > nbzero):
# Estimates the real start time of the IPU
ipustarttime = curtime - (len(ipupitch)) + 1
try:
# It is supposed ipupitch starts at time = 0.
iputargets = self.momel.annotate(ipupitch)
except Exception as e:
if self.logfile is not None:
self.logfile.print_message(
'No Momel annotation between time ' +
str(ipustarttime * 0.01) + " and " +
str(curtime * 0.01) +
" due to the following error: " + str(e),
indent=2,
status=-1)
else:
print("Momel Error: " + str(e))
iputargets = []
pass
# Adjust time values in the targets
for i in range(len(iputargets)):
x = iputargets[i].get_x()
iputargets[i].set_x(ipustarttime + x)
# add this targets to the targets list
targets = targets + iputargets
del ipupitch[:]
curtime += 1
# last ipu
iputargets = []
if len(ipupitch) > 0 and (len(ipupitch) > nbzero):
try:
iputargets = self.momel.annotate(ipupitch)
except Exception as e:
if self.logfile is not None:
self.logfile.print_message(
'No Momel annotation between time ' +
str(ipustarttime * 0.01) + " and " +
str(curtime * 0.01) + " due to the following error: " +
str(e),
indent=2,
status=-1)
else:
print("error: " + str(e))
iputargets = []
pass
ipustarttime = curtime - (len(ipupitch))
# Adjust time values
for i in range(len(iputargets)):
x = iputargets[i].get_x()
iputargets[i].set_x(ipustarttime + x)
targets = targets + iputargets
# Print results and/or estimate INTSINT (if any)
if trsoutput:
trsm = Transcription("TrsMomel")
if outputfile:
momeltier = self.print_targets(targets, outputfile, trs=trsm)
else:
momeltier = self.print_targets(targets,
output_filename=None,
trs=trsm)
if self.logfile is not None:
self.logfile.print_message(str(len(targets)) +
" targets found.",
indent=2,
status=3)
momeltier.SetRadius(
0.005) # because one pitch estimation each 10ms...
sppas.src.annotationdata.aio.write(trsoutput, trsm)
elif outputfile:
self.print_targets(targets, outputfile, trs=None)
else:
self.print_targets(targets, output_filename='STDOUT', trs=None)
# ----------------------------------------------------------------------------
# Read
trsinput = sppas.src.annotationdata.aio.read(args.i)
# Take all tiers or specified tiers
tiersnumbs = list()
if not args.t:
tiersnumbs = range(1, (trsinput.GetSize() + 1))
elif args.t:
tiersnumbs = args.t
# ----------------------------------------------------------------------------
# Fill
trsout = Transcription()
for i in tiersnumbs:
tier = trsinput[i - 1]
tier = fill_gaps(tier, trsinput.GetMinTime(), trsinput.GetMaxTime())
ctrlvocab = tier.GetCtrlVocab()
if ctrlvocab is not None:
if ctrlvocab.Contains(args.f) is False:
ctrlvocab.Append(args.f, descr="Filler")
print "Tier: ", tier.GetName()
print "Fill empty intervals with", args.f, "(and merge with previous or following if any)"
tier = fct_fill(tier, args.f)
print "Merge intervals during less than", args.d
tier = fct_clean(tier, args.f, args.d)
def create_chunks(self, inputaudio, phontier, toktier, diralign):
""" Create time-aligned tiers from raw intput tiers.
:param inputaudio: (str) Name of the audio file
:param phontier: (Tier) the tier with phonetization
:param toktier: (Tier) the tier with tokenization to split
:param diralign: (str) the working directory to store temporary data.
"""
trsoutput = Transcription("Chunks")
# Extract the audio channel
channel = autils.extract_audio_channel(inputaudio, 0)
channel = autils.format_channel(channel, 16000, 2)
# Extract the lists of tokens and their corresponding pronunciations
pronlist = self._tier2raw(phontier, map=True).split()
toklist = self._tier2raw(toktier, map=False).split()
if len(pronlist) != len(toklist):
raise IOError("Inconsistency between the number of items in "
"phonetization %d and tokenization %d." % (len(pronlist), len(toklist)))
# At a first stage, we'll find anchors.
anchor_tier = AnchorTier()
anchor_tier.set_duration(channel.get_duration())
anchor_tier.set_ext_delay(1.)
anchor_tier.set_out_delay(0.5)
# Search silences and use them as anchors.
if self.SILENCES is True:
anchor_tier.append_silences(channel)
# Estimates the speaking rate (amount of tokens/sec. in average)
self._spkrate.eval_from_duration(channel.get_duration(), len(toklist))
# Multi-pass ASR to find anchors
nb_anchors = -1 # number of anchors in the preceding pass
ngram = self.N # decreasing N-gram value
win_length = self.W # decreasing window length
while nb_anchors != anchor_tier.GetSize() and anchor_tier.check_holes_ntokens(self.NBT) is False:
anchor_tier.set_win_delay(win_length)
nb_anchors = anchor_tier.GetSize()
logging.debug(" =========================================================== ")
logging.debug(" Number of anchors: %d" % nb_anchors)
logging.debug(" N-gram: %d" % ngram)
logging.debug(" W-length: %d" % win_length)
# perform ASR and append new anchors in the anchor tier (if any)
self._asr(toklist, pronlist, anchor_tier, channel, diralign, ngram)
# append the anchor tier as intermediate result
if self.ANCHORS is True and nb_anchors != anchor_tier.GetSize():
Chunks._append_tier(anchor_tier, trsoutput)
out_name = os.path.join(diralign, "ANCHORS-%d.xra" % anchor_tier.GetSize())
sppas.src.annotationdata.aio.write(out_name, trsoutput)
# prepare next pass
win_length = max(win_length-1., self.WMIN)
ngram = max(ngram-1, self.NMIN)
# Then, anchors are exported as tracks.
tiert = anchor_tier.export(toklist)
tiert.SetName("Chunks-Tokenized")
tierp = anchor_tier.export(pronlist)
tierp.SetName("Chunks-Phonetized")
trsoutput.Append(tiert)
trsoutput.Append(tierp)
return trsoutput
if len(sys.argv) <= 1:
sys.argv.append('-h')
args = parser.parse_args()
if not args.quiet:
setup_logging(0, None)
else:
setup_logging(30, None)
# ----------------------------------------------------------------------------
# Read
trs_input = sppas.src.annotationdata.aio.read(args.i)
trs_out = Transcription()
# ----------------------------------------------------------------------------
# Transform the PhonAlign tier to a Phonetization tier
try:
align_tier = sppasSearchTier.aligned_phones(trs_input)
logging.info("PhonAlign tier found.")
phon_tier = unalign(align_tier)
phon_tier.SetName("Phones")
trs_out.Add(phon_tier)
except IOError:
logging.info("PhonAlign tier not found.")
# ----------------------------------------------------------------------------
# Transform the TokensAlign tier to a Tokenization tier
if fname.endswith("-palign") is False:
print("ERROR: MarsaTag plugin requires SPPAS alignment files "
"(i.e. with -palign in its name).")
sys.exit(1)
if fext.lower() != "textgrid":
# read content
trs_input = aio.read(filename)
tier = trs_input.Find("TokensAlign", case_sensitive=False)
if tier is None:
print("ERROR: A tier with name TokensAlign is required.")
sys.exit(1)
# write as textgrid
trs = Transcription(name="TokensAlign")
trs.Append(tier)
filename = fname + ".TextGrid"
aio.write(filename, trs)
# ----------------------------------------------------------------------------
# Get MarsaTag path
if len(args.p) == 0:
print("ERROR: No given directory for MarsaTag software tool.")
sys.exit(1)
if os.path.isdir(args.p) is False:
print("ERROR: {:s} is not a valid directory.".format(args.p))
sys.exit(1)
if not args.quiet:
if len(phones) != len(mappings):
sys.stdout.write("%s (ignored) " % phoneme.encode('utf8'))
else:
sys.stdout.write("%s " % phoneme.encode('utf8'))
for name, value in zip(tiernames, phones):
mappings[name].add(phoneme, value)
if not args.quiet:
print("\ndone...")
# ----------------------------------------------------------------------------
# Convert input file
trs = Transcription(name="PhonemesClassification")
if not args.quiet:
print("Classifying...")
for name in mappings.keys():
if not args.quiet:
print(" - {:s}".format(name))
new_tier = mappings[name].map_tier(tier)
new_tier.SetName(name)
trs.Append(new_tier)
print("done...")
# ----------------------------------------------------------------------------
# Write converted tiers
if not args.quiet:
class IPUsTrs(object):
"""
:author: Brigitte Bigi
:organization: Laboratoire Parole et Langage, Aix-en-Provence, France
:contact: [email protected]
:license: GPL, v3
:copyright: Copyright (C) 2011-2017 Brigitte Bigi
:summary: An IPUs segmentation from an already annotated data file.
"""
def __init__(self, trs):
""" Creates a new IPUsTrs instance.
:param trs: (Transcription) Input transcription from which it's
possible to extract IPUs.
Expected tiers are:
- first tier: the IPUs content [required]
- second tier: the IPUs file names [optional]
"""
super(IPUsTrs, self).__init__()
self._trsinput = Transcription()
self._units = list() # List of the content of the units (if any)
self._names = list() # List of file names for IPUs (if any)
self.set_transcription(trs)
# ------------------------------------------------------------------
def get_units(self):
""" Return the list of the IPUs contents. """
return self._units
# ------------------------------------------------------------------
def get_names(self):
""" Return the list of file names for IPUs. """
return self._names
# ------------------------------------------------------------------
# Manage Transcription
# ------------------------------------------------------------------
def set_transcription(self, trs):
""" Set a new Transcription.
:param trs: (Transcription) Input transcription from which it's
possible to extract IPUs.
"""
if trs is not None:
self._trsinput = trs
else:
self._trsinput = Transcription()
# ------------------------------------------------------------------
# Units search
# ------------------------------------------------------------------
def extract_bounds(self):
""" Return bound values.
Bound values are boolean to know if we expect a silence at start
or end of the given transcription. It is relevant only if the
transcription was created from a non-aligned file.
"""
# False means that I DON'T know if there is a silence:
# It does not mean that there IS NOT a silence.
# However, True means that there is a silence, for sure!
bound_start = False
bound_end = False
if len(self._trsinput) > 0:
# Check tier
tier = self._trsinput[0]
if tier.GetSize() == 0:
raise IOError('Got no utterances.')
# Fix bounds
if tier[0].GetLabel().IsSilence() is True:
bound_start = True
if tier[-1].GetLabel().IsSilence() is True and tier.GetSize() > 1:
bound_end = True
return bound_start, bound_end
# ------------------------------------------------------------------
def extract(self):
""" Extract units and (if any) extract tracks and silences.
:returns: tracks and silences, with time as seconds.
"""
self._units = list()
self._names = list()
if self._trsinput.GetSize() == 0:
return [], []
trstier = self._trsinput[0]
nametier = None
if self._trsinput.GetSize() == 2:
nametier = self._trsinput[1]
tracks = []
silences = []
if trstier.GetSize() == 0:
raise IOError('Got no utterances.')
if trstier[0].GetLocation().GetValue().IsTimeInterval():
(tracks, silences) = self.extract_aligned(trstier, nametier)
else:
self.extract_units()
return tracks, silences
# ------------------------------------------------------------------
def extract_units(self):
""" Extract IPUs content from a non-aligned transcription file.
"""
self._units = []
self._names = []
tier = self._trsinput[0]
if tier.GetSize() == 0:
raise IOError('Got no utterances.')
i = 0
for ann in tier:
if ann.GetLabel().IsSilence() is False:
self._units.append(ann.GetLabel().GetValue())
self._names.append("track_%.06d" % (i+1))
i += 1
# ------------------------------------------------------------------
def extract_aligned(self, trstier, nametier):
""" Extract from a time-aligned transcription file.
:returns: a tuple with tracks and silences lists
"""
trstracks = []
silences = []
self._units = list()
self._names = list()
i = 0
last = trstier.GetSize()
while i < last:
# Set the current annotation values
ann = trstier[i]
# Save information
if ann.GetLabel().IsSilence():
start = ann.GetLocation().GetBegin().GetMidpoint()
end = ann.GetLocation().GetEnd().GetMidpoint()
# Verify next annotations (concatenate all silences between 2 tracks)
if (i + 1) < last:
nextann = trstier[i + 1]
while (i + 1) < last and nextann.GetLabel().IsSilence():
end = nextann.GetLocation().GetEnd().GetMidpoint()
i += 1
if (i + 1) < last:
nextann = trstier[i + 1]
silences.append([start, end])
else:
start = ann.GetLocation().GetBegin().GetMidpoint()
end = ann.GetLocation().GetEnd().GetMidpoint()
trstracks.append([start, end])
self._units.append(ann.GetLabel().GetValue())
if nametier is not None:
aname = nametier.Find(ann.GetLocation().GetBegin().GetMidpoint(),
ann.GetLocation().GetEnd().GetMidpoint(), True)
if len(aname) == 0:
trstracks.pop()
self._units.pop()
else:
sf = sppasFileUtils(aname[0].GetLabel().GetValue())
# We have to take care in case of duplicated names
filename = sf.clear_whitespace()
if len(filename) == 0:
filename = "unnamed_track"
new_name = filename
idx = 2
while new_name in self._names:
new_name = u"%s_%.06d" % (filename, idx)
idx += 1
self._names.append(new_name)
# Continue
i += 1
return trstracks, silences