Update mg.trial4 (1).py
Browse files- mg.trial4 (1).py +5 -6
mg.trial4 (1).py
CHANGED
@@ -20,7 +20,7 @@ _DATA_URL = {
|
|
20 |
}
|
21 |
|
22 |
_TEXT_URL = {
|
23 |
-
"test": _DATA_ARCHIVE_ROOT1 + "
|
24 |
|
25 |
|
26 |
class MGDB2Dataset(datasets.GeneratorBasedBuilder):
|
@@ -29,9 +29,9 @@ class MGDB2Dataset(datasets.GeneratorBasedBuilder):
|
|
29 |
description=_DESCRIPTION,
|
30 |
features=datasets.Features(
|
31 |
{
|
32 |
-
|
33 |
"audio": datasets.Audio(sampling_rate=16_000),
|
34 |
-
"
|
35 |
}
|
36 |
),
|
37 |
supervised_keys=None,
|
@@ -43,9 +43,8 @@ class MGDB2Dataset(datasets.GeneratorBasedBuilder):
|
|
43 |
def _split_generators(self, dl_manager):
|
44 |
wav_archive = dl_manager.download(_DATA_URL)
|
45 |
txt_archive = dl_manager.download(_TEXT_URL)
|
46 |
-
test_dir = "
|
47 |
-
|
48 |
-
train_dir = "dataset/train"
|
49 |
|
50 |
|
51 |
print("Starting write datasets.........................................................")
|
|
|
20 |
}
|
21 |
|
22 |
_TEXT_URL = {
|
23 |
+
"test": _DATA_ARCHIVE_ROOT1 + "text.tsv",
|
24 |
|
25 |
|
26 |
class MGDB2Dataset(datasets.GeneratorBasedBuilder):
|
|
|
29 |
description=_DESCRIPTION,
|
30 |
features=datasets.Features(
|
31 |
{
|
32 |
+
|
33 |
"audio": datasets.Audio(sampling_rate=16_000),
|
34 |
+
"sentence": datasets.Value("string"),
|
35 |
}
|
36 |
),
|
37 |
supervised_keys=None,
|
|
|
43 |
def _split_generators(self, dl_manager):
|
44 |
wav_archive = dl_manager.download(_DATA_URL)
|
45 |
txt_archive = dl_manager.download(_TEXT_URL)
|
46 |
+
test_dir = "audio/ar/test"
|
47 |
+
|
|
|
48 |
|
49 |
|
50 |
print("Starting write datasets.........................................................")
|