datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
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.silero_open_stt_opus
Description
only subset tts_russian_addresses_rhvoice_4voices
Usage
from datasets import load_dataset, Audio
dataset = load_dataset("Sh1man/silero_open_stt_opus", "tts_russian_addresses_rhvoice_4voices", split="train")
print(dataset[0]['opus'])
Subsets
The dataset contains three subsets:
tts_russian_addresses_rhvoice_4voices: address recordings
📊 Сводная статистика аудио-датасетов
Информация по сплитам
🔹… See the full description on the dataset page: https://huggingface.co/datasets/Sh1man/silero_open_stt_opus.