Datasets:
Commit
•
73ca4fe
1
Parent(s):
3aebd00
Delete loading script
Browse files- librispeech_asr.py +0 -279
librispeech_asr.py
DELETED
@@ -1,279 +0,0 @@
|
|
1 |
-
# coding=utf-8
|
2 |
-
# Copyright 2021 The TensorFlow Datasets Authors and the HuggingFace Datasets Authors.
|
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 |
-
|
16 |
-
# Lint as: python3
|
17 |
-
"""Librispeech automatic speech recognition dataset."""
|
18 |
-
|
19 |
-
|
20 |
-
import os
|
21 |
-
|
22 |
-
import datasets
|
23 |
-
from datasets.tasks import AutomaticSpeechRecognition
|
24 |
-
|
25 |
-
|
26 |
-
_CITATION = """\
|
27 |
-
@inproceedings{panayotov2015librispeech,
|
28 |
-
title={Librispeech: an ASR corpus based on public domain audio books},
|
29 |
-
author={Panayotov, Vassil and Chen, Guoguo and Povey, Daniel and Khudanpur, Sanjeev},
|
30 |
-
booktitle={Acoustics, Speech and Signal Processing (ICASSP), 2015 IEEE International Conference on},
|
31 |
-
pages={5206--5210},
|
32 |
-
year={2015},
|
33 |
-
organization={IEEE}
|
34 |
-
}
|
35 |
-
"""
|
36 |
-
|
37 |
-
_DESCRIPTION = """\
|
38 |
-
LibriSpeech is a corpus of approximately 1000 hours of read English speech with sampling rate of 16 kHz,
|
39 |
-
prepared by Vassil Panayotov with the assistance of Daniel Povey. The data is derived from read
|
40 |
-
audiobooks from the LibriVox project, and has been carefully segmented and aligned.87
|
41 |
-
"""
|
42 |
-
|
43 |
-
_URL = "http://www.openslr.org/12"
|
44 |
-
_DL_URL = "http://www.openslr.org/resources/12/"
|
45 |
-
|
46 |
-
|
47 |
-
_DL_URLS = {
|
48 |
-
"clean": {
|
49 |
-
"dev": _DL_URL + "dev-clean.tar.gz",
|
50 |
-
"test": _DL_URL + "test-clean.tar.gz",
|
51 |
-
"train.100": _DL_URL + "train-clean-100.tar.gz",
|
52 |
-
"train.360": _DL_URL + "train-clean-360.tar.gz",
|
53 |
-
},
|
54 |
-
"other": {
|
55 |
-
"test": _DL_URL + "test-other.tar.gz",
|
56 |
-
"dev": _DL_URL + "dev-other.tar.gz",
|
57 |
-
"train.500": _DL_URL + "train-other-500.tar.gz",
|
58 |
-
},
|
59 |
-
"all": {
|
60 |
-
"dev.clean": _DL_URL + "dev-clean.tar.gz",
|
61 |
-
"dev.other": _DL_URL + "dev-other.tar.gz",
|
62 |
-
"test.clean": _DL_URL + "test-clean.tar.gz",
|
63 |
-
"test.other": _DL_URL + "test-other.tar.gz",
|
64 |
-
"train.clean.100": _DL_URL + "train-clean-100.tar.gz",
|
65 |
-
"train.clean.360": _DL_URL + "train-clean-360.tar.gz",
|
66 |
-
"train.other.500": _DL_URL + "train-other-500.tar.gz",
|
67 |
-
},
|
68 |
-
}
|
69 |
-
|
70 |
-
|
71 |
-
class LibrispeechASRConfig(datasets.BuilderConfig):
|
72 |
-
"""BuilderConfig for LibriSpeechASR."""
|
73 |
-
|
74 |
-
def __init__(self, **kwargs):
|
75 |
-
"""
|
76 |
-
Args:
|
77 |
-
data_dir: `string`, the path to the folder containing the files in the
|
78 |
-
downloaded .tar
|
79 |
-
citation: `string`, citation for the data set
|
80 |
-
url: `string`, url for information about the data set
|
81 |
-
**kwargs: keyword arguments forwarded to super.
|
82 |
-
"""
|
83 |
-
super(LibrispeechASRConfig, self).__init__(version=datasets.Version("2.1.0", ""), **kwargs)
|
84 |
-
|
85 |
-
|
86 |
-
class LibrispeechASR(datasets.GeneratorBasedBuilder):
|
87 |
-
"""Librispeech dataset."""
|
88 |
-
|
89 |
-
DEFAULT_WRITER_BATCH_SIZE = 256
|
90 |
-
DEFAULT_CONFIG_NAME = "all"
|
91 |
-
BUILDER_CONFIGS = [
|
92 |
-
LibrispeechASRConfig(name="clean", description="'Clean' speech."),
|
93 |
-
LibrispeechASRConfig(name="other", description="'Other', more challenging, speech."),
|
94 |
-
LibrispeechASRConfig(name="all", description="Combined clean and other dataset."),
|
95 |
-
]
|
96 |
-
|
97 |
-
def _info(self):
|
98 |
-
return datasets.DatasetInfo(
|
99 |
-
description=_DESCRIPTION,
|
100 |
-
features=datasets.Features(
|
101 |
-
{
|
102 |
-
"file": datasets.Value("string"),
|
103 |
-
"audio": datasets.Audio(sampling_rate=16_000),
|
104 |
-
"text": datasets.Value("string"),
|
105 |
-
"speaker_id": datasets.Value("int64"),
|
106 |
-
"chapter_id": datasets.Value("int64"),
|
107 |
-
"id": datasets.Value("string"),
|
108 |
-
}
|
109 |
-
),
|
110 |
-
supervised_keys=("file", "text"),
|
111 |
-
homepage=_URL,
|
112 |
-
citation=_CITATION,
|
113 |
-
task_templates=[AutomaticSpeechRecognition(audio_column="audio", transcription_column="text")],
|
114 |
-
)
|
115 |
-
|
116 |
-
def _split_generators(self, dl_manager):
|
117 |
-
archive_path = dl_manager.download(_DL_URLS[self.config.name])
|
118 |
-
# (Optional) In non-streaming mode, we can extract the archive locally to have actual local audio files:
|
119 |
-
local_extracted_archive = dl_manager.extract(archive_path) if not dl_manager.is_streaming else {}
|
120 |
-
|
121 |
-
if self.config.name == "clean":
|
122 |
-
train_splits = [
|
123 |
-
datasets.SplitGenerator(
|
124 |
-
name="train.100",
|
125 |
-
gen_kwargs={
|
126 |
-
"local_extracted_archive": local_extracted_archive.get("train.100"),
|
127 |
-
"files": dl_manager.iter_archive(archive_path["train.100"]),
|
128 |
-
},
|
129 |
-
),
|
130 |
-
datasets.SplitGenerator(
|
131 |
-
name="train.360",
|
132 |
-
gen_kwargs={
|
133 |
-
"local_extracted_archive": local_extracted_archive.get("train.360"),
|
134 |
-
"files": dl_manager.iter_archive(archive_path["train.360"]),
|
135 |
-
},
|
136 |
-
),
|
137 |
-
]
|
138 |
-
dev_splits = [
|
139 |
-
datasets.SplitGenerator(
|
140 |
-
name=datasets.Split.VALIDATION,
|
141 |
-
gen_kwargs={
|
142 |
-
"local_extracted_archive": local_extracted_archive.get("dev"),
|
143 |
-
"files": dl_manager.iter_archive(archive_path["dev"]),
|
144 |
-
},
|
145 |
-
)
|
146 |
-
]
|
147 |
-
test_splits = [
|
148 |
-
datasets.SplitGenerator(
|
149 |
-
name=datasets.Split.TEST,
|
150 |
-
gen_kwargs={
|
151 |
-
"local_extracted_archive": local_extracted_archive.get("test"),
|
152 |
-
"files": dl_manager.iter_archive(archive_path["test"]),
|
153 |
-
},
|
154 |
-
)
|
155 |
-
]
|
156 |
-
elif self.config.name == "other":
|
157 |
-
train_splits = [
|
158 |
-
datasets.SplitGenerator(
|
159 |
-
name="train.500",
|
160 |
-
gen_kwargs={
|
161 |
-
"local_extracted_archive": local_extracted_archive.get("train.500"),
|
162 |
-
"files": dl_manager.iter_archive(archive_path["train.500"]),
|
163 |
-
},
|
164 |
-
)
|
165 |
-
]
|
166 |
-
dev_splits = [
|
167 |
-
datasets.SplitGenerator(
|
168 |
-
name=datasets.Split.VALIDATION,
|
169 |
-
gen_kwargs={
|
170 |
-
"local_extracted_archive": local_extracted_archive.get("dev"),
|
171 |
-
"files": dl_manager.iter_archive(archive_path["dev"]),
|
172 |
-
},
|
173 |
-
)
|
174 |
-
]
|
175 |
-
test_splits = [
|
176 |
-
datasets.SplitGenerator(
|
177 |
-
name=datasets.Split.TEST,
|
178 |
-
gen_kwargs={
|
179 |
-
"local_extracted_archive": local_extracted_archive.get("test"),
|
180 |
-
"files": dl_manager.iter_archive(archive_path["test"]),
|
181 |
-
},
|
182 |
-
)
|
183 |
-
]
|
184 |
-
elif self.config.name == "all":
|
185 |
-
train_splits = [
|
186 |
-
datasets.SplitGenerator(
|
187 |
-
name="train.clean.100",
|
188 |
-
gen_kwargs={
|
189 |
-
"local_extracted_archive": local_extracted_archive.get("train.clean.100"),
|
190 |
-
"files": dl_manager.iter_archive(archive_path["train.clean.100"]),
|
191 |
-
},
|
192 |
-
),
|
193 |
-
datasets.SplitGenerator(
|
194 |
-
name="train.clean.360",
|
195 |
-
gen_kwargs={
|
196 |
-
"local_extracted_archive": local_extracted_archive.get("train.clean.360"),
|
197 |
-
"files": dl_manager.iter_archive(archive_path["train.clean.360"]),
|
198 |
-
},
|
199 |
-
),
|
200 |
-
datasets.SplitGenerator(
|
201 |
-
name="train.other.500",
|
202 |
-
gen_kwargs={
|
203 |
-
"local_extracted_archive": local_extracted_archive.get("train.other.500"),
|
204 |
-
"files": dl_manager.iter_archive(archive_path["train.other.500"]),
|
205 |
-
},
|
206 |
-
),
|
207 |
-
]
|
208 |
-
dev_splits = [
|
209 |
-
datasets.SplitGenerator(
|
210 |
-
name="validation.clean",
|
211 |
-
gen_kwargs={
|
212 |
-
"local_extracted_archive": local_extracted_archive.get("dev.clean"),
|
213 |
-
"files": dl_manager.iter_archive(archive_path["dev.clean"]),
|
214 |
-
},
|
215 |
-
),
|
216 |
-
datasets.SplitGenerator(
|
217 |
-
name="validation.other",
|
218 |
-
gen_kwargs={
|
219 |
-
"local_extracted_archive": local_extracted_archive.get("dev.other"),
|
220 |
-
"files": dl_manager.iter_archive(archive_path["dev.other"]),
|
221 |
-
},
|
222 |
-
),
|
223 |
-
]
|
224 |
-
test_splits = [
|
225 |
-
datasets.SplitGenerator(
|
226 |
-
name="test.clean",
|
227 |
-
gen_kwargs={
|
228 |
-
"local_extracted_archive": local_extracted_archive.get("test.clean"),
|
229 |
-
"files": dl_manager.iter_archive(archive_path["test.clean"]),
|
230 |
-
},
|
231 |
-
),
|
232 |
-
datasets.SplitGenerator(
|
233 |
-
name="test.other",
|
234 |
-
gen_kwargs={
|
235 |
-
"local_extracted_archive": local_extracted_archive.get("test.other"),
|
236 |
-
"files": dl_manager.iter_archive(archive_path["test.other"]),
|
237 |
-
},
|
238 |
-
),
|
239 |
-
]
|
240 |
-
|
241 |
-
return train_splits + dev_splits + test_splits
|
242 |
-
|
243 |
-
def _generate_examples(self, files, local_extracted_archive):
|
244 |
-
"""Generate examples from a LibriSpeech archive_path."""
|
245 |
-
key = 0
|
246 |
-
audio_data = {}
|
247 |
-
transcripts = []
|
248 |
-
for path, f in files:
|
249 |
-
if path.endswith(".flac"):
|
250 |
-
id_ = path.split("/")[-1][: -len(".flac")]
|
251 |
-
audio_data[id_] = f.read()
|
252 |
-
elif path.endswith(".trans.txt"):
|
253 |
-
for line in f:
|
254 |
-
if line:
|
255 |
-
line = line.decode("utf-8").strip()
|
256 |
-
id_, transcript = line.split(" ", 1)
|
257 |
-
audio_file = f"{id_}.flac"
|
258 |
-
speaker_id, chapter_id = [int(el) for el in id_.split("-")[:2]]
|
259 |
-
audio_file = (
|
260 |
-
os.path.join(local_extracted_archive, audio_file)
|
261 |
-
if local_extracted_archive
|
262 |
-
else audio_file
|
263 |
-
)
|
264 |
-
transcripts.append(
|
265 |
-
{
|
266 |
-
"id": id_,
|
267 |
-
"speaker_id": speaker_id,
|
268 |
-
"chapter_id": chapter_id,
|
269 |
-
"file": audio_file,
|
270 |
-
"text": transcript,
|
271 |
-
}
|
272 |
-
)
|
273 |
-
if audio_data and len(audio_data) == len(transcripts):
|
274 |
-
for transcript in transcripts:
|
275 |
-
audio = {"path": transcript["file"], "bytes": audio_data[transcript["id"]]}
|
276 |
-
yield key, {"audio": audio, **transcript}
|
277 |
-
key += 1
|
278 |
-
audio_data = {}
|
279 |
-
transcripts = []
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|