Hail may
Browse files- hinglish.py +18 -11
hinglish.py
CHANGED
@@ -3,14 +3,24 @@ import json
|
|
3 |
import os
|
4 |
import datasets
|
5 |
|
|
|
|
|
|
|
|
|
|
|
6 |
_ANNOT_URL = {
|
7 |
-
"train": "
|
8 |
-
"test": "
|
9 |
}
|
10 |
|
|
|
|
|
|
|
|
|
|
|
11 |
_DATA_URL = [
|
12 |
-
"
|
13 |
-
"
|
14 |
]
|
15 |
|
16 |
_DESCRIPTION = """\
|
@@ -28,21 +38,19 @@ class HinglishDataset(datasets.GeneratorBasedBuilder):
|
|
28 |
"audio": datasets.Audio(sampling_rate=16_000),
|
29 |
"sentence": datasets.Value("string"),
|
30 |
}),
|
31 |
-
supervised_keys=
|
32 |
)
|
33 |
|
34 |
def _split_generators(self, dl_manager):
|
35 |
prompts_paths = dl_manager.download(_ANNOT_URL)
|
36 |
archive = dl_manager.download(_DATA_URL)
|
37 |
-
|
38 |
-
test_dir = "hinglish/data/test"
|
39 |
-
|
40 |
return [
|
41 |
datasets.SplitGenerator(
|
42 |
name=datasets.Split.TRAIN,
|
43 |
gen_kwargs={
|
44 |
"prompts_path": prompts_paths["train"],
|
45 |
-
"path_to_clips":
|
46 |
"audio_files": dl_manager.iter_archive(archive),
|
47 |
},
|
48 |
),
|
@@ -50,7 +58,7 @@ class HinglishDataset(datasets.GeneratorBasedBuilder):
|
|
50 |
name=datasets.Split.TEST,
|
51 |
gen_kwargs={
|
52 |
"prompts_path": prompts_paths["test"],
|
53 |
-
"path_to_clips":
|
54 |
"audio_files": dl_manager.iter_archive(archive),
|
55 |
},
|
56 |
),
|
@@ -61,7 +69,6 @@ class HinglishDataset(datasets.GeneratorBasedBuilder):
|
|
61 |
with open(prompts_path, encoding="utf-8") as f:
|
62 |
for row in f:
|
63 |
data = row.strip().split(",")
|
64 |
-
print(data)
|
65 |
audio_path = "/".join([path_to_clips, data[0]])
|
66 |
examples[audio_path] = {
|
67 |
"path": audio_path,
|
|
|
3 |
import os
|
4 |
import datasets
|
5 |
|
6 |
+
# _ANNOT_URL = {
|
7 |
+
# "train": "https://huggingface.co/datasets/ujs/hinglish/resolve/main/data/metadata.csv",
|
8 |
+
# "test": "https://huggingface.co/datasets/ujs/hinglish/resolve/main/data/metadata-test.csv",
|
9 |
+
# }
|
10 |
+
|
11 |
_ANNOT_URL = {
|
12 |
+
"train": "./data/metadata.csv",
|
13 |
+
"test": "./data/metadata-test.csv"
|
14 |
}
|
15 |
|
16 |
+
# _DATA_URL = [
|
17 |
+
# "https://huggingface.co/datasets/ujs/hinglish/resolve/main/data/train.tar.gz",
|
18 |
+
# "https://huggingface.co/datasets/ujs/hinglish/resolve/main/data/test.tar.gz"
|
19 |
+
# ]
|
20 |
+
|
21 |
_DATA_URL = [
|
22 |
+
"./data/train.tar.gz",
|
23 |
+
"./data/test.tar.gz"
|
24 |
]
|
25 |
|
26 |
_DESCRIPTION = """\
|
|
|
38 |
"audio": datasets.Audio(sampling_rate=16_000),
|
39 |
"sentence": datasets.Value("string"),
|
40 |
}),
|
41 |
+
supervised_keys=None,
|
42 |
)
|
43 |
|
44 |
def _split_generators(self, dl_manager):
|
45 |
prompts_paths = dl_manager.download(_ANNOT_URL)
|
46 |
archive = dl_manager.download(_DATA_URL)
|
47 |
+
data_dir = './data'
|
|
|
|
|
48 |
return [
|
49 |
datasets.SplitGenerator(
|
50 |
name=datasets.Split.TRAIN,
|
51 |
gen_kwargs={
|
52 |
"prompts_path": prompts_paths["train"],
|
53 |
+
"path_to_clips": data_dir,
|
54 |
"audio_files": dl_manager.iter_archive(archive),
|
55 |
},
|
56 |
),
|
|
|
58 |
name=datasets.Split.TEST,
|
59 |
gen_kwargs={
|
60 |
"prompts_path": prompts_paths["test"],
|
61 |
+
"path_to_clips": data_dir,
|
62 |
"audio_files": dl_manager.iter_archive(archive),
|
63 |
},
|
64 |
),
|
|
|
69 |
with open(prompts_path, encoding="utf-8") as f:
|
70 |
for row in f:
|
71 |
data = row.strip().split(",")
|
|
|
72 |
audio_path = "/".join([path_to_clips, data[0]])
|
73 |
examples[audio_path] = {
|
74 |
"path": audio_path,
|