Datasets:
Convert dataset to Parquet (part 00001-of-00002)
#7
by
albertvillanova
HF staff
- opened
- LICENSE.txt +0 -11
- README.md +24 -0
- tv3_0.3_train.fileids → data/test-00000-of-00002.parquet +2 -2
- tv3_0.3_train.transcription → data/test-00001-of-00002.parquet +2 -2
- tv3_0.3_test.fileids → data/train-00050-of-00068.parquet +2 -2
- tv3_0.3_test.transcription → data/train-00051-of-00068.parquet +2 -2
- data/train-00052-of-00068.parquet +3 -0
- data/train-00053-of-00068.parquet +3 -0
- data/train-00054-of-00068.parquet +3 -0
- data/train-00055-of-00068.parquet +3 -0
- data/train-00056-of-00068.parquet +3 -0
- data/train-00057-of-00068.parquet +3 -0
- data/train-00058-of-00068.parquet +3 -0
- data/train-00059-of-00068.parquet +3 -0
- data/train-00060-of-00068.parquet +3 -0
- data/train-00061-of-00068.parquet +3 -0
- data/train-00062-of-00068.parquet +3 -0
- data/train-00063-of-00068.parquet +3 -0
- data/train-00064-of-00068.parquet +3 -0
- data/train-00065-of-00068.parquet +3 -0
- data/train-00066-of-00068.parquet +3 -0
- data/train-00067-of-00068.parquet +3 -0
- tv3_0.3.tar.gz +0 -3
- tv3_parla.py +0 -111
LICENSE.txt
DELETED
@@ -1,11 +0,0 @@
|
|
1 |
-
Copyright (c) 2018 Col·lectivaT SCCL
|
2 |
-
|
3 |
-
The TV3Parla speech corpus is distributed under a Creative Commons
|
4 |
-
Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) license.
|
5 |
-
For details see <https://creativecommons.org/licenses/by-nc/4.0/>
|
6 |
-
|
7 |
-
The audio files and transcriptions that are the basis of the corpus are the
|
8 |
-
property of the Corporació Catalana de Mitjans Audiovisuals, SA (CCMA). Any
|
9 |
-
derivative work needs to refer to Parlament de Catalunya as the original owner
|
10 |
-
of the content and confirm with their terms of use as explained in their
|
11 |
-
website. <http://www.ccma.cat/avis-legal/condicions-utilitzacio-del-portal/>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
README.md
CHANGED
@@ -19,6 +19,30 @@ task_categories:
|
|
19 |
task_ids:
|
20 |
- language-modeling
|
21 |
pretty_name: TV3Parla
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
22 |
---
|
23 |
|
24 |
# Dataset Card for TV3Parla
|
|
|
19 |
task_ids:
|
20 |
- language-modeling
|
21 |
pretty_name: TV3Parla
|
22 |
+
dataset_info:
|
23 |
+
features:
|
24 |
+
- name: path
|
25 |
+
dtype: string
|
26 |
+
- name: audio
|
27 |
+
dtype: audio
|
28 |
+
- name: text
|
29 |
+
dtype: string
|
30 |
+
splits:
|
31 |
+
- name: train
|
32 |
+
num_bytes: 33609299680.048
|
33 |
+
num_examples: 159242
|
34 |
+
- name: test
|
35 |
+
num_bytes: 627492346.34
|
36 |
+
num_examples: 2220
|
37 |
+
download_size: 34204460155
|
38 |
+
dataset_size: 34236792026.388
|
39 |
+
configs:
|
40 |
+
- config_name: default
|
41 |
+
data_files:
|
42 |
+
- split: train
|
43 |
+
path: data/train-*
|
44 |
+
- split: test
|
45 |
+
path: data/test-*
|
46 |
---
|
47 |
|
48 |
# Dataset Card for TV3Parla
|
tv3_0.3_train.fileids → data/test-00000-of-00002.parquet
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ea22bc68ca2514e33d8902f7037c915eaec3817ccbbd88fa1c5ab3899644d4fd
|
3 |
+
size 309644027
|
tv3_0.3_train.transcription → data/test-00001-of-00002.parquet
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:79cc9d6046f198378c91f0e775f0ef4aed0ba1009f3d686d748b7be04a810f23
|
3 |
+
size 316758463
|
tv3_0.3_test.fileids → data/train-00050-of-00068.parquet
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:47a62914d09b57538ab5fadf74730eefa5486fecd89c02912e8b7982b25a7b32
|
3 |
+
size 465865861
|
tv3_0.3_test.transcription → data/train-00051-of-00068.parquet
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:52b1da0499cd9368e32c9eab9c3f4877415e17d480e62d6597a03563dd00b31a
|
3 |
+
size 487350331
|
data/train-00052-of-00068.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a07df709d99fd46b62350bed6dac86910c26e1941c2d3643c17fe47ec0f949b8
|
3 |
+
size 469916299
|
data/train-00053-of-00068.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a805d00c23a6b61cb60eb833293716c3cef551830721133156e839c73ff436dc
|
3 |
+
size 529869624
|
data/train-00054-of-00068.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4134a6e091e5f24c15735b931fe5b1e9498f11dded8bda12ad5e526560f3a704
|
3 |
+
size 505609118
|
data/train-00055-of-00068.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:81dcef913df0dc26b960df4e975a683346996938d794fdbe8f330728700067f0
|
3 |
+
size 474281000
|
data/train-00056-of-00068.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d2cc1aabaf1ca774238093057dbca0dde4ff60d0ea49e3c23b92618a2325ce24
|
3 |
+
size 499822014
|
data/train-00057-of-00068.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c7f4e067c99638078c72c31975e89aef1b9d0c09e10e643e091cc6a891b545ee
|
3 |
+
size 490327853
|
data/train-00058-of-00068.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:57650392e7b2964e53bc0c63f7f0c2c071a446cdcbe027d2cc9f860e27ca7fcd
|
3 |
+
size 471084496
|
data/train-00059-of-00068.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2bc845b7dff0476fb1eaa1956b58c48aaa62a91a18a257bd7f2dffc836d6ed60
|
3 |
+
size 507802197
|
data/train-00060-of-00068.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:88493daa80064ac182680698114e79665f229f5bde53fc11a7a6844496fdeb6f
|
3 |
+
size 523353945
|
data/train-00061-of-00068.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f73e203f7e49bbb806c5a03670a5a6066d7801bd68d24118b563b1ca8a9a46ab
|
3 |
+
size 502754596
|
data/train-00062-of-00068.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:881cba42d5000d05705e1de067fbb0aa2d1801d5fd8c3a87786534a78f9671eb
|
3 |
+
size 500159465
|
data/train-00063-of-00068.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4f2587373f94467726d2dbd220f74914cf0310964047bec289ca775dd1d1a820
|
3 |
+
size 549987985
|
data/train-00064-of-00068.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:80caa57bde9a5b05e70ac023090d53e2e91a8c3c30800c056cbaddd8575fb3ab
|
3 |
+
size 536379259
|
data/train-00065-of-00068.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:28d83e5835251e074d4aed9ad3d04aa6ac0de9625e72ea4f4af6ad8576d9f7c9
|
3 |
+
size 477341893
|
data/train-00066-of-00068.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:93557909e825f5e89918d4b57005bd5a525e4332502bc37883b635088af2d428
|
3 |
+
size 528730998
|
data/train-00067-of-00068.parquet
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5857d0cf2290fd4d8919e2a46682fa07027460113f06a8c22576bbc3b8cf73a8
|
3 |
+
size 559668574
|
tv3_0.3.tar.gz
DELETED
@@ -1,3 +0,0 @@
|
|
1 |
-
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:ccdd9a0eadadf53731fd2f33b3a24c1d300f3899fbd9419e707065214432298d
|
3 |
-
size 29637597121
|
|
|
|
|
|
|
|
tv3_parla.py
DELETED
@@ -1,111 +0,0 @@
|
|
1 |
-
# coding=utf-8
|
2 |
-
# Copyright 2020 The HuggingFace Datasets Authors and the current dataset script contributor.
|
3 |
-
#
|
4 |
-
# Licensed under the Apache License, Version 2.0 (the "License");
|
5 |
-
# you may not use this file except in compliance with the License.
|
6 |
-
# You may obtain a copy of the License at
|
7 |
-
#
|
8 |
-
# http://www.apache.org/licenses/LICENSE-2.0
|
9 |
-
#
|
10 |
-
# Unless required by applicable law or agreed to in writing, software
|
11 |
-
# distributed under the License is distributed on an "AS IS" BASIS,
|
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 |
-
"""TV3Parla."""
|
16 |
-
|
17 |
-
import re
|
18 |
-
|
19 |
-
import datasets
|
20 |
-
from datasets.tasks import AutomaticSpeechRecognition
|
21 |
-
|
22 |
-
|
23 |
-
_CITATION = """\
|
24 |
-
@inproceedings{kulebi18_iberspeech,
|
25 |
-
author={Baybars Külebi and Alp Öktem},
|
26 |
-
title={{Building an Open Source Automatic Speech Recognition System for Catalan}},
|
27 |
-
year=2018,
|
28 |
-
booktitle={Proc. IberSPEECH 2018},
|
29 |
-
pages={25--29},
|
30 |
-
doi={10.21437/IberSPEECH.2018-6}
|
31 |
-
}
|
32 |
-
"""
|
33 |
-
|
34 |
-
_DESCRIPTION = """\
|
35 |
-
This corpus includes 240 hours of Catalan speech from broadcast material.
|
36 |
-
The details of segmentation, data processing and also model training are explained in Külebi, Öktem; 2018.
|
37 |
-
The content is owned by Corporació Catalana de Mitjans Audiovisuals, SA (CCMA);
|
38 |
-
we processed their material and hereby making it available under their terms of use.
|
39 |
-
|
40 |
-
This project was supported by the Softcatalà Association.
|
41 |
-
"""
|
42 |
-
|
43 |
-
_HOMEPAGE = "https://collectivat.cat/asr#tv3parla"
|
44 |
-
|
45 |
-
_LICENSE = "Creative Commons Attribution-NonCommercial 4.0 International"
|
46 |
-
|
47 |
-
_REPO = "https://huggingface.co/datasets/collectivat/tv3_parla/resolve/main/"
|
48 |
-
_URLS = {
|
49 |
-
"transcripts": _REPO + "tv3_0.3_{split}.transcription",
|
50 |
-
"audio": _REPO + "tv3_0.3.tar.gz",
|
51 |
-
}
|
52 |
-
_SPLITS = [datasets.Split.TRAIN, datasets.Split.TEST]
|
53 |
-
|
54 |
-
_PATTERN = re.compile(r"^<s> (?P<text>.+) </s> \((?P<id>\S+)\)$")
|
55 |
-
|
56 |
-
|
57 |
-
class Tv3Parla(datasets.GeneratorBasedBuilder):
|
58 |
-
"""TV3Parla."""
|
59 |
-
|
60 |
-
VERSION = datasets.Version("0.3.0")
|
61 |
-
|
62 |
-
def _info(self):
|
63 |
-
return datasets.DatasetInfo(
|
64 |
-
description=_DESCRIPTION,
|
65 |
-
features=datasets.Features(
|
66 |
-
{
|
67 |
-
"path": datasets.Value("string"),
|
68 |
-
"audio": datasets.features.Audio(),
|
69 |
-
"text": datasets.Value("string"),
|
70 |
-
}
|
71 |
-
),
|
72 |
-
supervised_keys=None,
|
73 |
-
homepage=_HOMEPAGE,
|
74 |
-
license=_LICENSE,
|
75 |
-
citation=_CITATION,
|
76 |
-
task_templates=[
|
77 |
-
AutomaticSpeechRecognition(transcription_column="text")
|
78 |
-
],
|
79 |
-
)
|
80 |
-
|
81 |
-
def _split_generators(self, dl_manager):
|
82 |
-
urls = {
|
83 |
-
split: {key: url.format(split=split) for key, url in _URLS.items()} for split in _SPLITS
|
84 |
-
}
|
85 |
-
dl_dir = dl_manager.download(urls)
|
86 |
-
return [
|
87 |
-
datasets.SplitGenerator(
|
88 |
-
name=split,
|
89 |
-
gen_kwargs={
|
90 |
-
"transcripts_path": dl_dir[split]["transcripts"],
|
91 |
-
"audio_files": dl_manager.iter_archive(dl_dir[split]["audio"]),
|
92 |
-
"split": split,
|
93 |
-
},
|
94 |
-
) for split in _SPLITS
|
95 |
-
]
|
96 |
-
|
97 |
-
def _generate_examples(self, transcripts_path, audio_files, split):
|
98 |
-
transcripts = {}
|
99 |
-
with open(transcripts_path, encoding="utf-8") as transcripts_file:
|
100 |
-
for line in transcripts_file:
|
101 |
-
match = _PATTERN.match(line)
|
102 |
-
transcripts[match["id"]] = match["text"]
|
103 |
-
# train: 159242; test: 2220
|
104 |
-
for key, (path, file) in enumerate(audio_files):
|
105 |
-
if path.endswith(".wav") and f"/{split}/" in path:
|
106 |
-
uid = path.split("/")[-1][:-4]
|
107 |
-
if uid not in transcripts:
|
108 |
-
continue
|
109 |
-
text = transcripts.pop(uid)
|
110 |
-
audio = {"path": path, "bytes": file.read()}
|
111 |
-
yield key, {"path": path, "audio": audio, "text": text}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|