datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
laions_got_talent
LAION's Got Talent: Generated Voice Acting Dataset
Overview
"LAION's Got Talent" is a generated dataset comprising voice acting samples that exhibit a wide range of emotions, vocal bursts, topics, and content. This dataset is a component of the BUD-E project, spearheaded by LAION with support from Intel.
Dataset Composition
The dataset includes:
Emotional Diversity: Samples portraying various emotions to facilitate research in emotional recognition and… See the full description on the dataset page: https://huggingface.co/datasets/laion/laions_got_talent.laions_got_talent_rawlaions_got_talent_enhanced_no_metadatalaions_got_talent_german_bicodecTaiwan-Tongues-ASR-CE-dataset-hokkien
Taiwan-Tongues-ASR-CE-dataset-hokkien
本資料集為 Taiwan-Tongues-ASR-CE 專案所使用的預訓練資料，透過 WebDataset 格式打包，並上傳至 Hugging Face 以便研究人員與開發者自由取用。
📂 Dataset 結構
本資料集分為 Training 與 Test 兩個子集，均以 WebDataset tar 檔案形式存放：
Training set (WebDataset format)
train/train-000000.tar
train/train-000001.tar
...
Test set (WebDataset format)
test/test-000000.tar
...
tsv set
train.tsv
test.tsv
...
每個 tar 內部均包含對應的音檔與標註，方便直接搭配 WebDataset 與 PyTorch / Hugging Face datasets 進行訓練與測試。… See the full description on the dataset page: https://huggingface.co/datasets/adi-gov-tw/Taiwan-Tongues-ASR-CE-dataset-hokkien.laions_got_talent_previewTaiwan-Tongues-ASR-CE-dataset-zhtw
Taiwan-Tongues-ASR-CE-dataset-zhtw
本資料集為 Taiwan-Tongues-ASR-CE 專案所使用的預訓練資料，透過 WebDataset 格式打包，並上傳至 Hugging Face 以便研究人員與開發者自由取用。
📂 Dataset 結構
本資料集分為 Training 與 Test 兩個子集，均以 WebDataset tar 檔案形式存放：
Training set (WebDataset format)
train/train-000000.tar
train/train-000001.tar
...
Test set (WebDataset format)
test/test-000000.tar
...
tsv set
train.tsv
test.tsv
...
每個 tar 內部均包含對應的音檔與標註，方便直接搭配 WebDataset 與 PyTorch / Hugging Face datasets 進行訓練與測試。
🏷️… See the full description on the dataset page: https://huggingface.co/datasets/adi-gov-tw/Taiwan-Tongues-ASR-CE-dataset-zhtw.Taiwan-Tongues-ASR-CE-dataset-hakka
Taiwan-Tongues-ASR-CE-dataset-hakka
本資料集為 Taiwan-Tongues-ASR-CE 專案所使用的預訓練資料，透過 WebDataset 格式打包，並上傳至 Hugging Face 以便研究人員與開發者自由取用。
📂 Dataset 結構
本資料集分為 Training 與 Test 兩個子集，均以 WebDataset tar 檔案形式存放：
Training set (WebDataset format)
train/train-000000.tar
train/train-000001.tar
...
Test set (WebDataset format)
test/test-000000.tar
...
tsv set
train.tsv
test.tsv
...
每個 tar 內部均包含對應的音檔與標註，方便直接搭配 WebDataset 與 PyTorch / Hugging Face datasets 進行訓練與測試。
🏷️… See the full description on the dataset page: https://huggingface.co/datasets/adi-gov-tw/Taiwan-Tongues-ASR-CE-dataset-hakka.golos_opus
Dataset Description
GOLOS is a Russian dataset for speech research. This is the OPUS version of the GOLOS dataset.
Usage
from datasets import load_dataset, Audio
dataset = load_dataset("Sh1man/golos_opus", "crowd", split="train")
print(dataset[0]['opus'])
Dataset Statistics
Dataset structure
Domain
Train files
Train hours
Test files
Test hours
Crowd
979 796
1 095
9 994
11.2
Farfield
124 003
132.4
1 916
1.4
Total
1 103 799
1… See the full description on the dataset page: https://huggingface.co/datasets/Sh1man/golos_opus.Taiwan-Tongues-ASR-CE-dataset-en
Taiwan-Tongues-ASR-CE-dataset-en
本資料集為 Taiwan-Tongues-ASR-CE 專案所使用的預訓練資料，透過 WebDataset 格式打包，並上傳至 Hugging Face 以便研究人員與開發者自由取用。
📂 Dataset 結構
本資料集分為 Training 與 Test 兩個子集，均以 WebDataset tar 檔案形式存放：
Training set (WebDataset format)
train/train-000000.tar
train/train-000001.tar
...
Test set (WebDataset format)
test/test-000000.tar
...
tsv set
train.tsv
test.tsv
...
每個 tar 內部均包含對應的音檔與標註，方便直接搭配 WebDataset 與 PyTorch / Hugging Face datasets 進行訓練與測試。
🏷️… See the full description on the dataset page: https://huggingface.co/datasets/adi-gov-tw/Taiwan-Tongues-ASR-CE-dataset-en.google_myanmar_asr_voices
Google Myanmar ASR Dataset (WebDataset Version)
This repository provides a clean, user-friendly, and robust version of the Google Myanmar ASR Dataset, which is derived from the OpenSLR-80 Burmese Speech Corpus.
This version has been carefully re-processed into the WebDataset format. Each sample consists of a .wav audio file and a clean .json metadata file, packaged into sharded .tar archives. This format is highly efficient for large-scale training of ASR models.… See the full description on the dataset page: https://huggingface.co/datasets/freococo/google_myanmar_asr_voices.laions-got-talent-thinkinggol-cache-v2
GOL feature cache v2 — audit and reconstruction guide
This audit covers gated repository otoha-project/gol-cache-v2 revision
2a370b14f94e16175d55d2afdd013e1257bae123. The repository has no card, schema,
producer version, license, or reconstruction instructions.
Repository structure
The cache is one tar stream split into 18 raw parts:
gol_cache.tar.part000 through part016: 21,474,836,480 bytes each;
gol_cache.tar.part017: 7,535,349,760 bytes;
total: 372,607,569… See the full description on the dataset page: https://huggingface.co/datasets/otoha-project/gol-cache-v2.laions-got-talentfiltered-gol-dataset
Filtered GOL Dataset
midralab/gol-dataset をTTS（Text-to-Speech）学習用にフィルタリングしたデータセットです。
データセット概要
項目
値
総再生時間
約1,880時間
サンプル数
約120万
話者数
380人
データサイズ
約280GB
形式
WebDataset (.tar)
音声形式
FLAC (44.1kHz, モノラル)
フィルタリング条件
基本フィルタ
テキスト長: 3文字以上
音声長: 1秒以上、60秒未満
話者フィルタ
話者あたり5時間以上の音声データを持つ話者のみ
テキストフィルタ（除外対象）
非言語テキスト（句読点のみ、空白のみなど）
顔文字 (^_^), (T_T) など
笑い表現 (笑), 文末の www
絵文字
英数字のみのテキスト
同一文字4回以上の繰り返し
データ構造… See the full description on the dataset page: https://huggingface.co/datasets/tts-dataset/filtered-gol-dataset.scotus-neil_gorsuch-audio
SCOTUS-sim audio: neil_gorsuch
Per-utterance audio clips from Oyez oral-argument mp3s, sliced at
the start_time / stop_time timestamps stored in the companion
scotus-sim/scotus-neil_gorsuch-training dataset.
Alignment
clip_NNNNN.wav in the tarball corresponds exactly to
audio_segments.jsonl[NNNNN] in the training companion dataset.
In metadata.jsonl each row carries the same 0-padded index in idx.
This supersedes the v1 tarball, which had systematic audio↔segment
index… See the full description on the dataset page: https://huggingface.co/datasets/scotus-sim/scotus-neil_gorsuch-audio.laions_got_talent
LAION's Got Talent: Generated Voice Acting Dataset
Overview
"LAION's Got Talent" is a generated dataset comprising voice acting samples that exhibit a wide range of emotions, vocal bursts, topics, and content. This dataset is a component of the BUD-E project, spearheaded by LAION with support from Intel.
Dataset Composition
The dataset includes:
Emotional Diversity: Samples portraying various emotions to facilitate research in emotional recognition and… See the full description on the dataset page: https://huggingface.co/datasets/zuhri025/laions_got_talent.