テドリウム
コレクションでコンテンツを整理
必要に応じて、コンテンツの保存と分類を行います。
TED-LIUM コーパスは英語の TED トークで、文字起こしがあり、16kHz でサンプリングされています。約118時間のスピーチが含まれています。
FeaturesDict({
'gender': ClassLabel(shape=(), dtype=int64, num_classes=3),
'id': string,
'speaker_id': string,
'speech': Audio(shape=(None,), dtype=int64),
'text': Text(shape=(), dtype=string),
})
特徴 | クラス | 形 | Dtype | 説明 |
---|
| 特徴辞書 | | | |
性別 | クラスラベル | | int64 | |
ID | テンソル | | 弦 | |
Speaker_id | テンソル | | 弦 | |
スピーチ | オーディオ | (なし、) | int64 | |
文章 | 文章 | | 弦 | |
@inproceedings{rousseau2012tedlium,
title={TED-LIUM: an Automatic Speech Recognition dedicated corpus},
author={Rousseau, Anthony and Del{\'e}glise, Paul and Est{\`e}ve, Yannick},
booktitle={Conference on Language Resources and Evaluation (LREC)},
pages={125--129},
year={2012}
}
tedlium/release1 (デフォルト設定)
構成の説明: TED-LIUM コーパスは英語の TED トークで、文字起こしがあり、16kHz でサンプリングされています。約118時間のスピーチが含まれています。
This is the TED-LIUM corpus release 1,
licensed under Creative Commons BY-NC-ND 3.0
(http://creativecommons.org/licenses/by-nc-nd/3.0/deed.en).
ホームページ: https://www.openslr.org/7/
ダウンロードサイズ: 19.82 GiB
データセットサイズ: 39.23 GiB
スプリット:
スプリット | 例 |
---|
'test' | 1,469 |
'train' | 56,803 |
'validation' | 591 |
テドリウム/リリース2
構成の説明: これは TED-LIUM コーパス リリース 2 であり、Creative Commons BY-NC-ND 3.0 ( http://creativecommons.org/licenses/by-nc-nd/3.0/deed.en ) の下でライセンスされています。
All talks and text are property of TED Conferences LLC.
The TED-LIUM corpus was made from audio talks and their transcriptions
available on the TED website. We have prepared and filtered these data
in order to train acoustic models to participate to the International
Workshop on Spoken Language Translation 2011 (the LIUM English/French
SLT system reached the first rank in the SLT task).
Contains 1495 talks and transcripts.
ホームページ: https://www.openslr.org/19/
ダウンロードサイズ: 34.26 GiB
データセットサイズ: 67.04 GiB
スプリット:
スプリット | 例 |
---|
'test' | 1,469 |
'train' | 92,973 |
'validation' | 591 |
テドリウム/リリース3
構成の説明: これは TED-LIUM コーパス リリース 3 で、Creative Commons BY-NC-ND 3.0 の下でライセンスされています。
All talks and text are property of TED Conferences LLC.
This new TED-LIUM release was made through a collaboration between the
Ubiqus company and the LIUM (University of Le Mans, France)
Contents:
- 2351 audio talks in NIST sphere format (SPH), including talks from
TED-LIUM 2: be careful, same talks but not same audio files (only
these audio file must be used with the TED-LIUM 3 STM files)
- 452 hours of audio
- 2351 aligned automatic transcripts in STM format
- TEDLIUM 2 dev and test data: 19 TED talks in SPH format with
corresponding manual transcriptions (cf. 'legacy' distribution below).
- Dictionary with pronunciations (159848 entries), same file as the one
included in TED-LIUM 2
- Selected monolingual data for language modeling from WMT12 publicly
available corpora: these files come from the TED-LIUM 2 release, but
have been modified to get a tokenization more relevant for English
language
Two corpus distributions:
- the legacy one, on which the dev and test datasets are the same as in
TED-LIUM 2 (and TED-LIUM 1).
- the 'speaker adaptation' one, especially designed for experiments on
speaker adaptation.
ホームページ: https://www.openslr.org/51/
ダウンロードサイズ: 50.59 GiB
データセットサイズ: 145.67 GiB
スプリット:
スプリット | 例 |
---|
'test' | 1,469 |
'train' | 268,263 |
'validation' | 591 |
特に記載のない限り、このページのコンテンツはクリエイティブ・コモンズの表示 4.0 ライセンスにより使用許諾されます。コードサンプルは Apache 2.0 ライセンスにより使用許諾されます。詳しくは、Google Developers サイトのポリシーをご覧ください。Java は Oracle および関連会社の登録商標です。
最終更新日 2023-02-11 UTC。
[null,null,["最終更新日 2023-02-11 UTC。"],[],[],null,["# tedlium\n\n\u003cbr /\u003e\n\n- **Description**:\n\nThe TED-LIUM corpus is English-language TED talks, with transcriptions, sampled\nat 16kHz. It contains about 118 hours of speech.\n\n- **Additional Documentation** :\n [Explore on Papers With Code\n north_east](https://paperswithcode.com/dataset/ted-lium-3)\n\n- **Source code** :\n [`tfds.datasets.tedlium.Builder`](https://github.com/tensorflow/datasets/tree/master/tensorflow_datasets/datasets/tedlium/tedlium_dataset_builder.py)\n\n- **Versions**:\n\n - **`1.0.1`** (default): No release notes.\n- **Auto-cached**\n ([documentation](https://www.tensorflow.org/datasets/performances#auto-caching)):\n No\n\n- **Feature structure**:\n\n FeaturesDict({\n 'gender': ClassLabel(shape=(), dtype=int64, num_classes=3),\n 'id': string,\n 'speaker_id': string,\n 'speech': Audio(shape=(None,), dtype=int64),\n 'text': Text(shape=(), dtype=string),\n })\n\n- **Feature documentation**:\n\n| Feature | Class | Shape | Dtype | Description |\n|------------|--------------|---------|--------|-------------|\n| | FeaturesDict | | | |\n| gender | ClassLabel | | int64 | |\n| id | Tensor | | string | |\n| speaker_id | Tensor | | string | |\n| speech | Audio | (None,) | int64 | |\n| text | Text | | string | |\n\n- **Supervised keys** (See\n [`as_supervised` doc](https://www.tensorflow.org/datasets/api_docs/python/tfds/load#args)):\n `('speech', 'text')`\n\n- **Figure**\n ([tfds.show_examples](https://www.tensorflow.org/datasets/api_docs/python/tfds/visualization/show_examples)):\n Not supported.\n\n- **Citation**:\n\n @inproceedings{rousseau2012tedlium,\n title={TED-LIUM: an Automatic Speech Recognition dedicated corpus},\n author={Rousseau, Anthony and Del{\\'e}glise, Paul and Est{\\`e}ve, Yannick},\n booktitle={Conference on Language Resources and Evaluation (LREC)},\n pages={125--129},\n year={2012}\n }\n\ntedlium/release1 (default config)\n---------------------------------\n\n- **Config description**: The TED-LIUM corpus is English-language TED talks,\n with transcriptions, sampled at 16kHz. It contains about 118 hours of\n speech.\n\n This is the TED-LIUM corpus release 1,\n licensed under Creative Commons BY-NC-ND 3.0\n (http://creativecommons.org/licenses/by-nc-nd/3.0/deed.en).\n\n- **Homepage** : \u003chttps://www.openslr.org/7/\u003e\n\n- **Download size** : `19.82 GiB`\n\n- **Dataset size** : `39.23 GiB`\n\n- **Splits**:\n\n| Split | Examples |\n|----------------|----------|\n| `'test'` | 1,469 |\n| `'train'` | 56,803 |\n| `'validation'` | 591 |\n\n- **Examples** ([tfds.as_dataframe](https://www.tensorflow.org/datasets/api_docs/python/tfds/as_dataframe)):\n\nDisplay examples... \n\ntedlium/release2\n----------------\n\n- **Config description** : This is the TED-LIUM corpus release 2, licensed\n under Creative Commons BY-NC-ND 3.0\n (\u003chttp://creativecommons.org/licenses/by-nc-nd/3.0/deed.en\u003e).\n\n All talks and text are property of TED Conferences LLC.\n\n The TED-LIUM corpus was made from audio talks and their transcriptions\n available on the TED website. We have prepared and filtered these data\n in order to train acoustic models to participate to the International\n Workshop on Spoken Language Translation 2011 (the LIUM English/French\n SLT system reached the first rank in the SLT task).\n\n Contains 1495 talks and transcripts.\n\n- **Homepage** : \u003chttps://www.openslr.org/19/\u003e\n\n- **Download size** : `34.26 GiB`\n\n- **Dataset size** : `67.04 GiB`\n\n- **Splits**:\n\n| Split | Examples |\n|----------------|----------|\n| `'test'` | 1,469 |\n| `'train'` | 92,973 |\n| `'validation'` | 591 |\n\n- **Examples** ([tfds.as_dataframe](https://www.tensorflow.org/datasets/api_docs/python/tfds/as_dataframe)):\n\nDisplay examples... \n\ntedlium/release3\n----------------\n\n- **Config description**: This is the TED-LIUM corpus release 3, licensed\n under Creative Commons BY-NC-ND 3.0.\n\n All talks and text are property of TED Conferences LLC.\n\n This new TED-LIUM release was made through a collaboration between the\n Ubiqus company and the LIUM (University of Le Mans, France)\n\n Contents:\n\n - 2351 audio talks in NIST sphere format (SPH), including talks from\n TED-LIUM 2: be careful, same talks but not same audio files (only\n these audio file must be used with the TED-LIUM 3 STM files)\n - 452 hours of audio\n - 2351 aligned automatic transcripts in STM format\n - TEDLIUM 2 dev and test data: 19 TED talks in SPH format with\n corresponding manual transcriptions (cf. 'legacy' distribution below).\n - Dictionary with pronunciations (159848 entries), same file as the one\n included in TED-LIUM 2\n - Selected monolingual data for language modeling from WMT12 publicly\n available corpora: these files come from the TED-LIUM 2 release, but\n have been modified to get a tokenization more relevant for English\n language\n\n Two corpus distributions:\n - the legacy one, on which the dev and test datasets are the same as in\n TED-LIUM 2 (and TED-LIUM 1).\n - the 'speaker adaptation' one, especially designed for experiments on\n speaker adaptation.\n\n- **Homepage** : \u003chttps://www.openslr.org/51/\u003e\n\n- **Download size** : `50.59 GiB`\n\n- **Dataset size** : `145.67 GiB`\n\n- **Splits**:\n\n| Split | Examples |\n|----------------|----------|\n| `'test'` | 1,469 |\n| `'train'` | 268,263 |\n| `'validation'` | 591 |\n\n- **Examples** ([tfds.as_dataframe](https://www.tensorflow.org/datasets/api_docs/python/tfds/as_dataframe)):\n\nDisplay examples..."]]