def __load__(self):
if super(TextGridTierResource, self).__load__() < 0:
return -1
if self.__tiernames__ is None:
print "No tiername has been set."
print "No TierFrame can be created."
return -1
if self.__cached_object__ is None:
self.__cached_object__ = open_intervalframe_from_textgrid(self.get_filepath())[self.__tiernames__]
return 0
def setUp(self):
self.dict1 = open_intervalframe_from_textgrid('data/r1-20120704-cam1-'
'head-sk'
'.TextGrid', 'utf-8')
self.dict2 = open_intervalframe_from_textgrid('data/r1-20120704-cam1-'
'head-zm.TextGrid',
'utf-8')
self.dict3 = open_intervalframe_from_textgrid('data/mytextgrid'
'.TextGrid', 'utf-8')
self.dict4 = dict()
self.dict4['speaker A'] = self.dict3['speaker A'][0:4].copy(deep=True)
self.dict4['speaker B'] = self.dict3['speaker B'][0:4].copy(deep=True)
shift_tiers(self.dict4, 1)
self.tier1 = self.dict1['HeadSK']
self.tier2 = self.dict2['HeadZM']
self.stream = open_streamframe_from_xiofile('data/fseeksmaller.xio.gz',
"lab-labtop/irioKinect 2",
window_size=5,
with_fields=None,
without_fields=None,
discard_duplicates=True,
start_time=0,
end_time=500,
timestamp_offset=None)
self.ifr = create_intervalframe_from_streamframe(self.stream,
'soundAngle',
lambda x: True \
if x <- 0.7323895 \
else False,
10)
self.iframe = open_intervalframe_from_textgrid('data/mytextgrid.Text'
'Grid', 'utf-8')
self.speaker_a = self.iframe['speaker A']
self.speaker_b = self.iframe['speaker B']
self.overlap_ab_shouldbe = [{'end_time': 3.47134775993, 'start_time':
3.4540353413, 'text': u'exactly and at '
'the very end of the corridor is just '
'the bathroom/I see yes'},
{'end_time': 3.8781895978200001, 'start_'
'time': 3.7223778301200001,
'text': u'exactly and at the very end of '
'the corridor is just the bathroom/yes'},
{'end_time': 7.3163672878900003,
'start_time': 6.9095254500000003,
'text': u'that you quasi such have so '
'zack zack zack/yes'},
{'end_time': 7.5933659860300002,
'start_time': 7.4029293810599999,
'text': u'that you quasi such have so '
'zack zack zack/yes'},
{'end_time': 11.168380433799999,
'start_time': 10.8740693171,
'text': u'know you ?/yes I think I'},
{'end_time': 11.644471946299999,
'start_time': 11.2895673643,
'text': u'if this the corridor is here a '
'room there a room there a room '
'there and above/yes I think I'},
{'end_time': 14.892213031400001,
'start_time': 13.1073713208,
'text': u'if this the corridor is here a '
'room there a room there a room there '
'and above/there a room there there and '
'then there yes'},
{'end_time': 15.558741148799999,
'start_time': 15.376960753100001,
'text': u'that is perfect/yes'},
{'end_time': 17.679512431399999,
'start_time': 17.4544509891,
'text': u'yes/sure'},
{'end_time': 27.8435647606,
'start_time': 27.5492536438,
'text': u'twentyfive/so I would already '
'gladly a large room have'}]
self.right_overlap_ab = pd.DataFrame(self.overlap_ab_shouldbe,
columns=['start_time', 'end_time',
'text'])
self.union_dict = [{'end_time': 2.5191647350899999, 'start_time': 0.0,
'text': 'union'},
{'end_time': 6.1304666115000002, 'start_time': \
2.80481964254,
'text': 'union'},
{'end_time': 9.5669817102499994,
'start_time': 6.7796823102500001,
'text': 'union'},
{'end_time': 15.0393685898, 'start_time': \
10.7528823866, 'text': 'union'},
{'end_time': 16.1646758009,
'start_time': 15.281742450599999,
'text': 'union'},
{'end_time': 17.3159516401,
'start_time': 17.064921569900001,
'text': 'union'},
{'end_time': 17.714137268599998,
'start_time': 17.4457947798,
'text': 'union'},
{'end_time': 19.393441876099999,
'start_time': 17.826667989699999,
'text': 'union'}]
self.union_mustbe = pd.DataFrame(self.union_dict, columns=['start_'
'time', 'end_time', 'text'])
self.invert = [{'start_time': 0., 'end_time': 547.016,
'text': 'nod'},
{'start_time': 547.464, 'end_time': 549.507,
'text': 'nod/turn-aw'},
{'start_time': 549.988, 'end_time': 556.808,
'text': 'turn-aw/nod-2'},
{'start_time': 557.404, 'end_time': 561.428,
'text': 'nod-2/nod-3'},
{'start_time': 563.345, 'end_time': 600.000,
'text': 'nod-3'}]
self.inversion = pd.DataFrame(self.invert, columns=['start_'
'time', 'end_time', 'text'])
self.tier1_inverted = [{'end_time': 547.01599999999996,
'start_time': 0.0, 'text': u'nod'},
{'end_time': 549.50699999999995,
'start_time': 547.46400000000006,
'text': u'nod/turn-aw'},
{'end_time': 556.80799999999999,
'start_time': 549.98800000000006,
'text': u'turn-aw/nod-2'},
{'end_time': 561.428, 'start_time': 557.404,
'text': u'nod-2/nod-3'},
{'end_time': 567.51099999999997,
'start_time': 563.34500000000003,
'text': u'nod-3/slide-right'},
{'end_time': 578.17100000000005,
'start_time': 568.02300000000002,
'text': u'slide-right/nod-4'},
{'end_time': 586.03599999999994,
'start_time': 579.53499999999997,
'text': u'nod-4/nod-3'},
{'end_time': 604.73699999999997,
'start_time': 587.10699999999997,
'text': u'nod-3/nod-4-turn-aw-tw'},
{'end_time': 609.11800000000005,
'start_time': 607.35699999999997,
'text': u'nod-4-turn-aw-tw/nod-4'},
{'end_time': 617.30899999999997,
'start_time': 611.16600000000005,
'text': u'nod-4/nod-5'}]
self.invert_tier1 = pd.DataFrame(self.tier1_inverted,
columns=['start_time', 'end_time', 'text'])
self.empty = pd.DataFrame([])
self.withna = [{'start_time': 3, 'end_time': 6, 'text': 'hi'},
{'start_time': 7, 'end_time': 8.5},
{'start_time': 9, 'end_time': 13, 'text': 'bye'}]
self.emptyint = pd.DataFrame(self.withna, columns=['start_time',
'end_time', 'text'])
self.emptyshouldbe = [{'end_time': 3.0, 'start_time': 0, 'text': 'hi'},
{'end_time': 7.0, 'start_time': 6,
'text': 'hi/empty'},
{'end_time': 9.0, 'start_time': 8.5,
'text': 'empty/bye'}]
self.emptyshouldbepd = pd.DataFrame(self.emptyshouldbe, columns= \
['start_time', 'end_time', 'text'])
self.no_overlap1 = [{'start_time': 3, 'end_time': 5, 'text': 'eins'},
{'start_time': 9, 'end_time': 15, 'text': 'zwei'},
{'start_time': 19, 'end_time': 21, 'text': 'drei'}]
self.no_overlap2 = [{'start_time': 0, 'end_time': 2.5, 'text': 'vier'},
{'start_time': 6, 'end_time': 7, 'text': 'fuenf'},
{'start_time': 23, 'end_time': 27, 'text': 'sechs'}]
self.pd_no_overlap1 = pd.DataFrame(self.no_overlap1,
columns=['start_time', 'end_time', 'text'])
self.pd_no_overlap2 = pd.DataFrame(self.no_overlap2,
columns=['start_time', 'end_time', 'text'])
self.convert = [{'start_time': 4, 'end_time': 5, 'text': 'eins'},
{'start_time': 9, 'end_time': 15, 'text': 'zwei'},
{'start_time': 19, 'end_time': 21, 'text': 'drei'}]
self.convert_pd = pd.DataFrame(self.convert,
columns=['start_time', 'end_time', 'text'])
convert_times_of_tier(self.convert_pd, lambda y: int(1000 * y))
self.converted = [{'start_time': 4000, 'end_time': 5000,
'text': 'eins'}, {'start_time': 9000,
'end_time': 15000, 'text': 'zwei'},
{'start_time': 19000, 'end_time': 21000,
'text': 'drei'}]
self.converted_pd = pd.DataFrame(self.converted,
columns=['start_time', 'end_time', 'text'])
self.tiers_shifted_a = [{'end_time': 4.47134775993,
'start_time': 3.80481964254,
'text': u'I see yes'},
{'end_time':4.8781895978200001,
'start_time': 4.7223778301200001,
'text': u'yes'},
{'end_time': 8.3163672878900003,
'start_time': 7.9095254500000003,
'text': u'yes'},
{'end_time': 8.5933659860300002,
'start_time': 8.4029293810599999,
'text': u'yes'}]
self.tiers_shifted_pd = pd.DataFrame(self.tiers_shifted_a,
columns=['start_time', 'end_time', 'text'])
self.shift_dict = [{'start_time': 3, 'end_time': 5, 'text': 'eins'},
{'start_time': 9, 'end_time': 15, 'text': 'zwei'},
{'start_time': 19, 'end_time': 21, 'text': 'drei'}]
self.shift = pd.DataFrame(self.shift_dict,
columns=['start_time', 'end_time', 'text'])
shift_tier(self.shift, 1)
self.shifted_dict = [{'start_time': 4, 'end_time': 6, 'text': 'eins'},
{'start_time': 10, 'end_time': 16, 'text': 'zwei'},
{'start_time': 20, 'end_time': 22, 'text': 'drei'}]
self.shifted = pd.DataFrame(self.shifted_dict,
columns=['start_time', 'end_time', 'text'])
for col in self.dict4['speaker A'].columns:
self.dict4['speaker A'].loc[:, col] = \
self.dict4['speaker A'][col].map(lambda x: str(x))
for col in self.tiers_shifted_pd.columns:
self.tiers_shifted_pd.loc[:, col] = \
self.tiers_shifted_pd[col].map(lambda x: str(x))
self.withPoint = open_intervalframe_from_textgrid('data/r1_12_15with'
'Point.TextGrid',
'utf-8')
self.points = self.withPoint['P'][0:4].copy(deep=True)
shift_tier(self.points, 10)
for col in self.points.columns:
self.points.loc[:, col] = self.points[col].map(lambda x: str(x))
self.points_shifted = [{'mark': 'A', 'time': 12.804819642542952},
{'mark': 'B', 'time': 13.454035341299345},
{'mark': 'A', 'time': 13.722377830118717},
{'mark': 'B', 'time': 16.779682310252952}]
self.points_shifted_pd = pd.DataFrame(self.points_shifted,
columns=['time', 'mark'])
for col in self.points_shifted_pd.columns:
self.points_shifted_pd.loc[:, col] = \
self.points_shifted_pd[col].map(lambda x: str(x))
self.labeljoin = open_intervalframe_from_textgrid('data/joinlabels'
'.TextGrid', 'utf-8')
self.streamdict1 = [{'value': u'nod_start'},
{'value': u'nod_end'},
{'value': u'turn-aw_start'},
{'value': u'turn-aw_end'},
{'value': u'nod-2_start'},
{'value': u'nod-2_end'},
{'value': u'nod-3_start'},
{'value': u'nod-3'},
{'value': u'nod-3_end'}]
self.stream1 = pd.DataFrame(self.streamdict1, columns=['value'])
self.stream1.index = [547.016, 547.464, 549.507, 549.988, 556.808,
557.404, 561.428, 562.428, 563.345]
self.inv1 = [{'end_time': '547.016', 'start_time': '300.0',
'text': u'nod'},
{'end_time': '549.507', 'start_time': '547.464',
'text': u'nod/turn-aw'},
{'end_time': '556.808', 'start_time': '549.988',
'text': u'turn-aw/nod-2'},
{'end_time': '561.428', 'start_time': '557.404',
'text': u'nod-2/nod-3'},
{'end_time': '700.0', 'start_time': '563.345',
'text': u'nod-3'}]
self.inverted1 = pd.DataFrame(self.inv1,
columns=['start_time', 'end_time', 'text'])
self.inv1_str = invert_intervalframe(self.tier1[0:4], 300, 700)
for col in self.inv1_str:
self.inv1_str.loc[:, col] = \
self.inv1_str[col].map(lambda x: str(x))
self.inv_default_conc = [{'end_time': '549.507',
'start_time': '547.464', 'text': u'nod'},
{'end_time': '556.808',
'start_time': '549.988', 'text': u'turn-aw+nod-2'},
{'end_time': '561.428',
'start_time': '557.404', 'text': u'nod-2'}]
self.inv_def_pd = pd.DataFrame(self.inv_default_conc,
columns=['start_time', 'end_time', 'text'])
self.inv_def_str = invert_intervalframe(self.tier1[0:4],
concat_delimiter='+')
for col in self.inv_def_str:
self.inv_def_str.loc[:, col] = \
self.inv_def_str[col].map(lambda x: str(x))
def setUp(self):
self.dict1 = open_intervalframe_from_textgrid(
'data/r1-20120704-cam1-'
'head-sk'
'.TextGrid', 'utf-8')
self.dict2 = open_intervalframe_from_textgrid(
'data/r1-20120704-cam1-'
'head-zm.TextGrid', 'utf-8')
self.dict3 = open_intervalframe_from_textgrid(
'data/mytextgrid'
'.TextGrid', 'utf-8')
self.dict4 = dict()
self.dict4['speaker A'] = self.dict3['speaker A'][0:4].copy(deep=True)
self.dict4['speaker B'] = self.dict3['speaker B'][0:4].copy(deep=True)
shift_tiers(self.dict4, 1)
self.tier1 = self.dict1['HeadSK']
self.tier2 = self.dict2['HeadZM']
self.stream = open_streamframe_from_xiofile('data/fseeksmaller.xio.gz',
"lab-labtop/irioKinect 2",
window_size=5,
with_fields=None,
without_fields=None,
discard_duplicates=True,
start_time=0,
end_time=500,
timestamp_offset=None)
self.ifr = create_intervalframe_from_streamframe(self.stream,
'soundAngle',
lambda x: True \
if x <- 0.7323895 \
else False,
10)
self.iframe = open_intervalframe_from_textgrid(
'data/mytextgrid.Text'
'Grid', 'utf-8')
self.speaker_a = self.iframe['speaker A']
self.speaker_b = self.iframe['speaker B']
self.overlap_ab_shouldbe = [{
'end_time':
3.47134775993,
'start_time':
3.4540353413,
'text':
u'exactly and at '
'the very end of the corridor is just '
'the bathroom/I see yes'
}, {
'end_time':
3.8781895978200001,
'start_'
'time':
3.7223778301200001,
'text':
u'exactly and at the very end of '
'the corridor is just the bathroom/yes'
}, {
'end_time':
7.3163672878900003,
'start_time':
6.9095254500000003,
'text':
u'that you quasi such have so '
'zack zack zack/yes'
}, {
'end_time':
7.5933659860300002,
'start_time':
7.4029293810599999,
'text':
u'that you quasi such have so '
'zack zack zack/yes'
}, {
'end_time': 11.168380433799999,
'start_time': 10.8740693171,
'text': u'know you ?/yes I think I'
}, {
'end_time':
11.644471946299999,
'start_time':
11.2895673643,
'text':
u'if this the corridor is here a '
'room there a room there a room '
'there and above/yes I think I'
}, {
'end_time':
14.892213031400001,
'start_time':
13.1073713208,
'text':
u'if this the corridor is here a '
'room there a room there a room there '
'and above/there a room there there and '
'then there yes'
}, {
'end_time': 15.558741148799999,
'start_time': 15.376960753100001,
'text': u'that is perfect/yes'
}, {
'end_time': 17.679512431399999,
'start_time': 17.4544509891,
'text': u'yes/sure'
}, {
'end_time':
27.8435647606,
'start_time':
27.5492536438,
'text':
u'twentyfive/so I would already '
'gladly a large room have'
}]
self.right_overlap_ab = pd.DataFrame(
self.overlap_ab_shouldbe,
columns=['start_time', 'end_time', 'text'])
self.union_dict = [{'end_time': 2.5191647350899999, 'start_time': 0.0,
'text': 'union'},
{'end_time': 6.1304666115000002, 'start_time': \
2.80481964254,
'text': 'union'},
{'end_time': 9.5669817102499994,
'start_time': 6.7796823102500001,
'text': 'union'},
{'end_time': 15.0393685898, 'start_time': \
10.7528823866, 'text': 'union'},
{'end_time': 16.1646758009,
'start_time': 15.281742450599999,
'text': 'union'},
{'end_time': 17.3159516401,
'start_time': 17.064921569900001,
'text': 'union'},
{'end_time': 17.714137268599998,
'start_time': 17.4457947798,
'text': 'union'},
{'end_time': 19.393441876099999,
'start_time': 17.826667989699999,
'text': 'union'}]
self.union_mustbe = pd.DataFrame(
self.union_dict, columns=['start_'
'time', 'end_time', 'text'])
self.invert = [{
'start_time': 0.,
'end_time': 547.016,
'text': 'nod'
}, {
'start_time': 547.464,
'end_time': 549.507,
'text': 'nod/turn-aw'
}, {
'start_time': 549.988,
'end_time': 556.808,
'text': 'turn-aw/nod-2'
}, {
'start_time': 557.404,
'end_time': 561.428,
'text': 'nod-2/nod-3'
}, {
'start_time': 563.345,
'end_time': 600.000,
'text': 'nod-3'
}]
self.inversion = pd.DataFrame(
self.invert, columns=['start_'
'time', 'end_time', 'text'])
self.tier1_inverted = [{
'end_time': 547.01599999999996,
'start_time': 0.0,
'text': u'nod'
}, {
'end_time': 549.50699999999995,
'start_time': 547.46400000000006,
'text': u'nod/turn-aw'
}, {
'end_time': 556.80799999999999,
'start_time': 549.98800000000006,
'text': u'turn-aw/nod-2'
}, {
'end_time': 561.428,
'start_time': 557.404,
'text': u'nod-2/nod-3'
}, {
'end_time': 567.51099999999997,
'start_time': 563.34500000000003,
'text': u'nod-3/slide-right'
}, {
'end_time': 578.17100000000005,
'start_time': 568.02300000000002,
'text': u'slide-right/nod-4'
}, {
'end_time': 586.03599999999994,
'start_time': 579.53499999999997,
'text': u'nod-4/nod-3'
}, {
'end_time': 604.73699999999997,
'start_time': 587.10699999999997,
'text': u'nod-3/nod-4-turn-aw-tw'
}, {
'end_time': 609.11800000000005,
'start_time': 607.35699999999997,
'text': u'nod-4-turn-aw-tw/nod-4'
}, {
'end_time': 617.30899999999997,
'start_time': 611.16600000000005,
'text': u'nod-4/nod-5'
}]
self.invert_tier1 = pd.DataFrame(
self.tier1_inverted, columns=['start_time', 'end_time', 'text'])
self.empty = pd.DataFrame([])
self.withna = [{
'start_time': 3,
'end_time': 6,
'text': 'hi'
}, {
'start_time': 7,
'end_time': 8.5
}, {
'start_time': 9,
'end_time': 13,
'text': 'bye'
}]
self.emptyint = pd.DataFrame(
self.withna, columns=['start_time', 'end_time', 'text'])
self.emptyshouldbe = [{
'end_time': 3.0,
'start_time': 0,
'text': 'hi'
}, {
'end_time': 7.0,
'start_time': 6,
'text': 'hi/empty'
}, {
'end_time': 9.0,
'start_time': 8.5,
'text': 'empty/bye'
}]
self.emptyshouldbepd = pd.DataFrame(self.emptyshouldbe, columns= \
['start_time', 'end_time', 'text'])
self.no_overlap1 = [{
'start_time': 3,
'end_time': 5,
'text': 'eins'
}, {
'start_time': 9,
'end_time': 15,
'text': 'zwei'
}, {
'start_time': 19,
'end_time': 21,
'text': 'drei'
}]
self.no_overlap2 = [{
'start_time': 0,
'end_time': 2.5,
'text': 'vier'
}, {
'start_time': 6,
'end_time': 7,
'text': 'fuenf'
}, {
'start_time': 23,
'end_time': 27,
'text': 'sechs'
}]
self.pd_no_overlap1 = pd.DataFrame(
self.no_overlap1, columns=['start_time', 'end_time', 'text'])
self.pd_no_overlap2 = pd.DataFrame(
self.no_overlap2, columns=['start_time', 'end_time', 'text'])
self.convert = [{
'start_time': 4,
'end_time': 5,
'text': 'eins'
}, {
'start_time': 9,
'end_time': 15,
'text': 'zwei'
}, {
'start_time': 19,
'end_time': 21,
'text': 'drei'
}]
self.convert_pd = pd.DataFrame(
self.convert, columns=['start_time', 'end_time', 'text'])
convert_times_of_tier(self.convert_pd, lambda y: int(1000 * y))
self.converted = [{
'start_time': 4000,
'end_time': 5000,
'text': 'eins'
}, {
'start_time': 9000,
'end_time': 15000,
'text': 'zwei'
}, {
'start_time': 19000,
'end_time': 21000,
'text': 'drei'
}]
self.converted_pd = pd.DataFrame(
self.converted, columns=['start_time', 'end_time', 'text'])
self.tiers_shifted_a = [{
'end_time': 4.47134775993,
'start_time': 3.80481964254,
'text': u'I see yes'
}, {
'end_time': 4.8781895978200001,
'start_time': 4.7223778301200001,
'text': u'yes'
}, {
'end_time': 8.3163672878900003,
'start_time': 7.9095254500000003,
'text': u'yes'
}, {
'end_time': 8.5933659860300002,
'start_time': 8.4029293810599999,
'text': u'yes'
}]
self.tiers_shifted_pd = pd.DataFrame(
self.tiers_shifted_a, columns=['start_time', 'end_time', 'text'])
self.shift_dict = [{
'start_time': 3,
'end_time': 5,
'text': 'eins'
}, {
'start_time': 9,
'end_time': 15,
'text': 'zwei'
}, {
'start_time': 19,
'end_time': 21,
'text': 'drei'
}]
self.shift = pd.DataFrame(self.shift_dict,
columns=['start_time', 'end_time', 'text'])
shift_tier(self.shift, 1)
self.shifted_dict = [{
'start_time': 4,
'end_time': 6,
'text': 'eins'
}, {
'start_time': 10,
'end_time': 16,
'text': 'zwei'
}, {
'start_time': 20,
'end_time': 22,
'text': 'drei'
}]
self.shifted = pd.DataFrame(self.shifted_dict,
columns=['start_time', 'end_time', 'text'])
for col in self.dict4['speaker A'].columns:
self.dict4['speaker A'].loc[:, col] = \
self.dict4['speaker A'][col].map(lambda x: str(x))
for col in self.tiers_shifted_pd.columns:
self.tiers_shifted_pd.loc[:, col] = \
self.tiers_shifted_pd[col].map(lambda x: str(x))
self.withPoint = open_intervalframe_from_textgrid(
'data/r1_12_15with'
'Point.TextGrid', 'utf-8')
self.points = self.withPoint['P'][0:4].copy(deep=True)
shift_tier(self.points, 10)
for col in self.points.columns:
self.points.loc[:, col] = self.points[col].map(lambda x: str(x))
self.points_shifted = [{
'mark': 'A',
'time': 12.804819642542952
}, {
'mark': 'B',
'time': 13.454035341299345
}, {
'mark': 'A',
'time': 13.722377830118717
}, {
'mark': 'B',
'time': 16.779682310252952
}]
self.points_shifted_pd = pd.DataFrame(self.points_shifted,
columns=['time', 'mark'])
for col in self.points_shifted_pd.columns:
self.points_shifted_pd.loc[:, col] = \
self.points_shifted_pd[col].map(lambda x: str(x))
self.labeljoin = open_intervalframe_from_textgrid(
'data/joinlabels'
'.TextGrid', 'utf-8')
self.streamdict1 = [{
'value': u'nod_start'
}, {
'value': u'nod_end'
}, {
'value': u'turn-aw_start'
}, {
'value': u'turn-aw_end'
}, {
'value': u'nod-2_start'
}, {
'value': u'nod-2_end'
}, {
'value': u'nod-3_start'
}, {
'value': u'nod-3'
}, {
'value': u'nod-3_end'
}]
self.stream1 = pd.DataFrame(self.streamdict1, columns=['value'])
self.stream1.index = [
547.016, 547.464, 549.507, 549.988, 556.808, 557.404, 561.428,
562.428, 563.345
]
self.inv1 = [{
'end_time': '547.016',
'start_time': '300.0',
'text': u'nod'
}, {
'end_time': '549.507',
'start_time': '547.464',
'text': u'nod/turn-aw'
}, {
'end_time': '556.808',
'start_time': '549.988',
'text': u'turn-aw/nod-2'
}, {
'end_time': '561.428',
'start_time': '557.404',
'text': u'nod-2/nod-3'
}, {
'end_time': '700.0',
'start_time': '563.345',
'text': u'nod-3'
}]
self.inverted1 = pd.DataFrame(
self.inv1, columns=['start_time', 'end_time', 'text'])
self.inv1_str = invert_intervalframe(self.tier1[0:4], 300, 700)
for col in self.inv1_str:
self.inv1_str.loc[:, col] = \
self.inv1_str[col].map(lambda x: str(x))
self.inv_default_conc = [{
'end_time': '549.507',
'start_time': '547.464',
'text': u'nod'
}, {
'end_time': '556.808',
'start_time': '549.988',
'text': u'turn-aw+nod-2'
}, {
'end_time': '561.428',
'start_time': '557.404',
'text': u'nod-2'
}]
self.inv_def_pd = pd.DataFrame(
self.inv_default_conc, columns=['start_time', 'end_time', 'text'])
self.inv_def_str = invert_intervalframe(self.tier1[0:4],
concat_delimiter='+')
for col in self.inv_def_str:
self.inv_def_str.loc[:, col] = \
self.inv_def_str[col].map(lambda x: str(x))
def setUp(self):
self.f = open_streamframe_from_xiofile(
"data/fseeksmaller.xio.gz",
"lab-labtop/irioKinect 2",
window_size=5,
with_fields=[],
without_fields=[],
discard_duplicates=True,
start_time=0,
end_time=13,
relative=True,
timestamp_offset=10,
)
self.f2 = open_streamframe_from_xiofile(
"data/fseeksmaller.xio.gz",
"lab-labtop/irioKinect",
window_size=5,
with_fields=[],
without_fields=[],
discard_duplicates=True,
start_time=0,
end_time=13,
relative=True,
timestamp_offset=10,
)
# self.outtake_from_stream = self.f.ix[1341393414826]['framenumber']
self.fraw = open_streamframe_from_xiofile(
"data/fseeksmaller.xio.gz",
"lab-labtop/irioKinect 2",
window_size=5,
with_fields=[],
without_fields=[],
discard_duplicates=True,
start_time=0,
end_time=13,
relative=True,
timestamp_offset="raw",
)
self.ff = open_streamframe_from_xiofile(
"data/fseeksmaller.xio.gz",
"lab-labtop/irioKinect 2",
window_size=5,
with_fields=[],
without_fields=[],
discard_duplicates=True,
start_time=0,
end_time=13,
relative=True,
)
save_streamframe_to_xiofile({"lab-labtop/irioKinect 2": self.f}, "data/sf_to_xio.xio.gz")
save_streamframe_to_xiofile(
{"lab-labtop/irioKinect 2": self.f, "lab-labtop/irioKinect": self.f2}, "data/sf_to_xio2.xio.gz"
)
self.rsn = open_streamframe_from_xiofile("data/fseeksmaller.xio.gz", "wrong/sensor/name")
self.outtake_from_xio = XIOFile("data/sf_to_xio.xio.gz", indexing=True)
self.outtake_from_xio_2 = XIOFile("data/sf_to_xio2.xio.gz", indexing=True)
self.q = [
ex for ex in quantize(self.outtake_from_xio.xio_quicklinegen(0, 13, True, True), "lab-labtop/irioKinect 2")
][0]["soundAngle"]
self.ivf = open_intervalframe_from_textgrid(
"data/r1_12_15with" "Point.TextGrid", encoding="utf-8", asobjects=False, include_empty_intervals=False
)["P"]
self.cv = convert_pointtier_to_streamframe(self.ivf)
self.pf = convert_streamframe_to_pointtier(self.f)
self.outtake_from_pf = convert_streamframe_to_pointtier(self.f)["soundAngle"].ix[0]["time"]
self.if_from_tg = open_intervalframe_from_textgrid(
"data/r1-20120704-" "cam1-head-zm.TextGrid",
encoding="utf-8",
asobjects=False,
include_empty_intervals=False,
)
self.if_from_tg_tier = self.if_from_tg.values()[0]
save_intervalframe_to_textgrid(self.if_from_tg, "data/testif.TextGrid", encoding="utf-8")
self.tg = tgt.read_textgrid("data/testif.TextGrid", encoding="utf-8", include_empty_intervals=False)
self.ic1 = open_intervalframe_from_increco("data/test.inc_reco")
self.ic2 = open_intervalframe_from_increco("data/test.inc_reco", lastonly=True)
ranges1 = [
line.strip() for line in open(
"/media/data/jh/simple_rnn_disf/rnn_disf_detection/data/disfluency_detection/swda_divisions_disfluency_detection/SWDisfHeldoutASR_ranges.text"
)
]
ranges2 = [
line.strip() for line in open(
"/media/data/jh/simple_rnn_disf/rnn_disf_detection/data/disfluency_detection/swda_divisions_disfluency_detection/SWDisfTestASR_ranges.text"
)
]
for ranges in [ranges1, ranges2]:
final_file.write("\n\n")
for r in ranges:
for s in ["A", "B"]:
iframe = open_intervalframe_from_textgrid("{0}{1}.TextGrid".format(
r, s))
hyp = " ".join(iframe['Hyp']['text'])
ref = " ".join(iframe['Ref']['text'])
wer = wer(ref, hyp)
cost = wer(ref, hyp, macro=True)
print r, s, wer
print >> final_file, r, s, wer, cost
final_file.close()
#Based on the results, output the 'good' ASR results
results = open("wer_test.text")
no_ho = 0
no_test = 0
ingood = True
file = open(
def setUp(self):
self.f = open_streamframe_from_xiofile('data/fseeksmaller.xio.gz',
"lab-labtop/irioKinect 2",
window_size=5, with_fields=[],
without_fields=[],
discard_duplicates=True,
start_time=0, end_time=13,
relative=True,
timestamp_offset=10)
self.f2 = open_streamframe_from_xiofile('data/fseeksmaller.xio.gz',
"lab-labtop/irioKinect",
window_size=5, with_fields=[],
without_fields=[],
discard_duplicates=True,
start_time=0, end_time=13,
relative=True,
timestamp_offset=10)
#self.outtake_from_stream = self.f.ix[1341393414826]['framenumber']
self.fraw = open_streamframe_from_xiofile('data/fseeksmaller.xio.gz',
"lab-labtop/irioKinect 2",
window_size=5, with_fields=[],
without_fields=[],
discard_duplicates=True,
start_time=0, end_time=13,
relative=True,
timestamp_offset='raw')
self.ff = open_streamframe_from_xiofile('data/fseeksmaller.xio.gz',
"lab-labtop/irioKinect 2",
window_size=5, with_fields=[],
without_fields=[],
discard_duplicates=True,
start_time=0, end_time=13,
relative=True)
save_streamframe_to_xiofile({"lab-labtop/irioKinect 2": self.f},
'data/sf_to_xio.xio.gz')
save_streamframe_to_xiofile({"lab-labtop/irioKinect 2": self.f,
"lab-labtop/irioKinect": self.f2},
'data/sf_to_xio2.xio.gz')
self.rsn = open_streamframe_from_xiofile('data/fseeksmaller.xio.gz',
"wrong/sensor/name")
self.outtake_from_xio = XIOFile('data/sf_to_xio.xio.gz',
indexing=True)
self.outtake_from_xio_2 = XIOFile('data/sf_to_xio2.xio.gz',
indexing=True)
self.q = [ex for ex in quantize(self.outtake_from_xio.xio_quicklinegen\
(0, 13, True, True),
"lab-labtop/irioKinect 2")][0]['soundAngle']
self.ivf = open_intervalframe_from_textgrid('data/r1_12_15with'
'Point.TextGrid',
encoding='utf-8',
asobjects = False,
include_empty_intervals = False)['P']
self.cv = convert_pointtier_to_streamframe(self.ivf)
self.pf = convert_streamframe_to_pointtier(self.f)
self.outtake_from_pf = \
convert_streamframe_to_pointtier(self.f)['soundAngle'].ix[0]['time']
self.if_from_tg = open_intervalframe_from_textgrid('data/r1-20120704-'
'cam1-head-zm.TextGrid',
encoding='utf-8',
asobjects = False,
include_empty_intervals = False)
self.if_from_tg_tier = self.if_from_tg.values()[0]
save_intervalframe_to_textgrid(self.if_from_tg,
'data/testif.TextGrid',
encoding='utf-8')
self.tg = tgt.read_textgrid('data/testif.TextGrid',
encoding='utf-8',
include_empty_intervals=False)
self.ic1 = open_intervalframe_from_increco('data/test.inc_reco')
self.ic2 = open_intervalframe_from_increco('data/test.inc_reco',
lastonly=True)
