Commit
·
00846d9
1
Parent(s):
a44f1ba
update
Browse files
evi.py
CHANGED
@@ -12,6 +12,12 @@
|
|
12 |
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
13 |
# See the License for the specific language governing permissions and
|
14 |
# limitations under the License.
|
|
|
|
|
|
|
|
|
|
|
|
|
15 |
|
16 |
import csv
|
17 |
import json
|
@@ -23,8 +29,6 @@ import datasets
|
|
23 |
logger = datasets.logging.get_logger(__name__)
|
24 |
|
25 |
|
26 |
-
""" EVI Dataset"""
|
27 |
-
|
28 |
_CITATION = """\
|
29 |
@inproceedings{Spithourakis2022evi,
|
30 |
author = {Georgios P. Spithourakis and Ivan Vuli\'{c} and Micha\l{} Lis and I\~{n}igo Casanueva and Pawe\l{} Budzianowski},
|
@@ -42,9 +46,15 @@ _ALL_CONFIGS = sorted([
|
|
42 |
|
43 |
_LANGS = sorted(["en", "fr", "pl"])
|
44 |
|
45 |
-
_DESCRIPTION = "
|
|
|
|
|
|
|
|
|
46 |
|
47 |
-
|
|
|
|
|
48 |
|
49 |
_BASE_URL = "https://huggingface.co/datasets/PolyAI/evi/resolve/main/data"
|
50 |
|
@@ -54,9 +64,7 @@ _TEXT_URL = {
|
|
54 |
|
55 |
_AUDIO_DATA_URL = "https://poly-public-data.s3.eu-west-2.amazonaws.com/evi-paper/audios.zip" # noqa
|
56 |
|
57 |
-
|
58 |
-
|
59 |
-
_VERSION = datasets.Version("0.0.10", "")
|
60 |
|
61 |
|
62 |
class EviConfig(datasets.BuilderConfig):
|
@@ -69,19 +77,10 @@ class EviConfig(datasets.BuilderConfig):
|
|
69 |
self.languages = _LANGS if name == "all" else [name.split("-")[0]] # all langs if config == "all"
|
70 |
|
71 |
|
72 |
-
def _build_config(name):
|
73 |
-
return EviConfig(
|
74 |
-
name=name,
|
75 |
-
version=_VERSION,
|
76 |
-
description=_DESCRIPTION,
|
77 |
-
# homepage=_HOMEPAGE_URL,
|
78 |
-
)
|
79 |
-
|
80 |
-
|
81 |
class Evi(datasets.GeneratorBasedBuilder):
|
82 |
|
83 |
-
DEFAULT_WRITER_BATCH_SIZE =
|
84 |
-
BUILDER_CONFIGS = [
|
85 |
|
86 |
def _info(self):
|
87 |
features = datasets.Features(
|
@@ -99,18 +98,16 @@ class Evi(datasets.GeneratorBasedBuilder):
|
|
99 |
)
|
100 |
|
101 |
return datasets.DatasetInfo(
|
102 |
-
version=
|
103 |
-
description=
|
104 |
-
license=
|
105 |
citation=_CITATION,
|
106 |
features=features,
|
|
|
107 |
)
|
108 |
|
109 |
def _split_generators(self, dl_manager):
|
110 |
-
|
111 |
langs = self.config.languages
|
112 |
-
# "./audios/{lang}/{dialogue_id}/3.wav"
|
113 |
-
|
114 |
lang2text_urls = {
|
115 |
lang: _TEXT_URL[lang] for lang in langs
|
116 |
}
|
@@ -121,10 +118,8 @@ class Evi(datasets.GeneratorBasedBuilder):
|
|
121 |
datasets.SplitGenerator(
|
122 |
name=datasets.Split.TEST,
|
123 |
gen_kwargs={
|
124 |
-
# "audio_data": dl_manager.iter_files(audio_data_path),
|
125 |
"audio_data_path": audio_data_path,
|
126 |
"text_paths": lang2text_paths,
|
127 |
-
# "streaming": dl_manager.is_streaming
|
128 |
},
|
129 |
)
|
130 |
]
|
@@ -137,9 +132,7 @@ class Evi(datasets.GeneratorBasedBuilder):
|
|
137 |
reader = csv.DictReader(
|
138 |
fin, delimiter="\t", skipinitialspace=True
|
139 |
)
|
140 |
-
for
|
141 |
-
if i == 199:
|
142 |
-
break
|
143 |
dialogue_id = dictrow["dialogue_id"]
|
144 |
turn_id = dictrow["turn_num"]
|
145 |
file_path = os.path.join(
|
@@ -148,7 +141,7 @@ class Evi(datasets.GeneratorBasedBuilder):
|
|
148 |
dialogue_id,
|
149 |
f'{turn_id}.wav'
|
150 |
)
|
151 |
-
full_path =
|
152 |
|
153 |
yield file_path, {
|
154 |
"language": lang,
|
|
|
12 |
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
13 |
# See the License for the specific language governing permissions and
|
14 |
# limitations under the License.
|
15 |
+
"""
|
16 |
+
EVI is a challenging spoken multilingual dataset with 5,506 dialogues in English, Polish, and French
|
17 |
+
that can be used for benchmarking and developing knowledge-based enrolment, identification, and identification
|
18 |
+
for spoken dialogue systems.
|
19 |
+
"""
|
20 |
+
|
21 |
|
22 |
import csv
|
23 |
import json
|
|
|
29 |
logger = datasets.logging.get_logger(__name__)
|
30 |
|
31 |
|
|
|
|
|
32 |
_CITATION = """\
|
33 |
@inproceedings{Spithourakis2022evi,
|
34 |
author = {Georgios P. Spithourakis and Ivan Vuli\'{c} and Micha\l{} Lis and I\~{n}igo Casanueva and Pawe\l{} Budzianowski},
|
|
|
46 |
|
47 |
_LANGS = sorted(["en", "fr", "pl"])
|
48 |
|
49 |
+
_DESCRIPTION = """
|
50 |
+
EVI is a challenging spoken multilingual dataset with 5,506 dialogues in English, Polish, and French
|
51 |
+
that can be used for benchmarking and developing knowledge-based enrolment, identification, and identification
|
52 |
+
for spoken dialogue systems.
|
53 |
+
""" # noqa
|
54 |
|
55 |
+
_LICENSE = "CC-BY-4.0"
|
56 |
+
|
57 |
+
_HOMEPAGE = "https://github.com/PolyAI-LDN/evi-paper"
|
58 |
|
59 |
_BASE_URL = "https://huggingface.co/datasets/PolyAI/evi/resolve/main/data"
|
60 |
|
|
|
64 |
|
65 |
_AUDIO_DATA_URL = "https://poly-public-data.s3.eu-west-2.amazonaws.com/evi-paper/audios.zip" # noqa
|
66 |
|
67 |
+
_VERSION = datasets.Version("0.0.1", "")
|
|
|
|
|
68 |
|
69 |
|
70 |
class EviConfig(datasets.BuilderConfig):
|
|
|
77 |
self.languages = _LANGS if name == "all" else [name.split("-")[0]] # all langs if config == "all"
|
78 |
|
79 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
80 |
class Evi(datasets.GeneratorBasedBuilder):
|
81 |
|
82 |
+
DEFAULT_WRITER_BATCH_SIZE = 512
|
83 |
+
BUILDER_CONFIGS = [EviConfig(name) for name in _ALL_CONFIGS + ["all"]]
|
84 |
|
85 |
def _info(self):
|
86 |
features = datasets.Features(
|
|
|
98 |
)
|
99 |
|
100 |
return datasets.DatasetInfo(
|
101 |
+
version=_VERSION,
|
102 |
+
description=_DESCRIPTION,
|
103 |
+
license=_LICENSE,
|
104 |
citation=_CITATION,
|
105 |
features=features,
|
106 |
+
homepage=_HOMEPAGE
|
107 |
)
|
108 |
|
109 |
def _split_generators(self, dl_manager):
|
|
|
110 |
langs = self.config.languages
|
|
|
|
|
111 |
lang2text_urls = {
|
112 |
lang: _TEXT_URL[lang] for lang in langs
|
113 |
}
|
|
|
118 |
datasets.SplitGenerator(
|
119 |
name=datasets.Split.TEST,
|
120 |
gen_kwargs={
|
|
|
121 |
"audio_data_path": audio_data_path,
|
122 |
"text_paths": lang2text_paths,
|
|
|
123 |
},
|
124 |
)
|
125 |
]
|
|
|
132 |
reader = csv.DictReader(
|
133 |
fin, delimiter="\t", skipinitialspace=True
|
134 |
)
|
135 |
+
for dictrow in reader:
|
|
|
|
|
136 |
dialogue_id = dictrow["dialogue_id"]
|
137 |
turn_id = dictrow["turn_num"]
|
138 |
file_path = os.path.join(
|
|
|
141 |
dialogue_id,
|
142 |
f'{turn_id}.wav'
|
143 |
)
|
144 |
+
full_path = os.path.join(audio_data_path, file_path)
|
145 |
|
146 |
yield file_path, {
|
147 |
"language": lang,
|