Ahmed-ibn-Harun
commited on
Commit
•
d0fcb79
1
Parent(s):
06aac9d
Update wake.py
Browse files
wake.py
CHANGED
@@ -42,7 +42,7 @@ _ALL_CONFIGS = sorted([
|
|
42 |
_DESCRIPTION = "Wake is a dataset for the wake word detection task with spoken data."
|
43 |
|
44 |
|
45 |
-
_DATA_URL =
|
46 |
|
47 |
|
48 |
class WakeConfig(datasets.BuilderConfig):
|
@@ -108,10 +108,10 @@ class Wake(datasets.GeneratorBasedBuilder):
|
|
108 |
|
109 |
archive_path = dl_manager.download_and_extract(self.config.data_url)
|
110 |
audio_path = dl_manager.extract(
|
111 |
-
os.path.join(archive_path,
|
112 |
)
|
113 |
text_path = dl_manager.extract(
|
114 |
-
os.path.join(archive_path, "
|
115 |
)
|
116 |
|
117 |
text_path = {l: os.path.join(text_path, f"{l}.csv") for l in langs}
|
|
|
42 |
_DESCRIPTION = "Wake is a dataset for the wake word detection task with spoken data."
|
43 |
|
44 |
|
45 |
+
_DATA_URL = 'https://huggingface.co/datasets/Ahmed-ibn-Harun/wake-w/resolve/main/data.tar.gz'
|
46 |
|
47 |
|
48 |
class WakeConfig(datasets.BuilderConfig):
|
|
|
108 |
|
109 |
archive_path = dl_manager.download_and_extract(self.config.data_url)
|
110 |
audio_path = dl_manager.extract(
|
111 |
+
os.path.join(archive_path, "audio.tar.gz")
|
112 |
)
|
113 |
text_path = dl_manager.extract(
|
114 |
+
os.path.join(archive_path, "text.tar.gz")
|
115 |
)
|
116 |
|
117 |
text_path = {l: os.path.join(text_path, f"{l}.csv") for l in langs}
|