metadata
description: >-
An utterance-level version of the Samples of Spoken Finnish corpus detailed
here: https://www.kielipankki.fi/corpora/skn/
If you use this dataset, refer to the original data providers: Institute for
the Languages of Finland (2021). Samples of Spoken Finnish, Downloadable
Version [data set]. Kielipankki. http://urn.fi/urn:nbn:fi:lb-2020112937
dataset_info:
features:
- name: 'Unnamed: 0'
dtype: int64
- name: orig_index
dtype: int64
- name: SpeakerID
dtype: string
- name: duration
dtype: float64
- name: original
dtype: string
- name: path
dtype: string
- name: DocID
dtype: string
- name: Location
dtype: string
- name: detailed
dtype: string
- name: simple
dtype: string
- name: split
dtype: string
- name: path_data
dtype:
audio:
sampling_rate: 16000
splits:
- name: train
num_bytes: 26152519293.768
num_examples: 42738
download_size: 25790159339
dataset_size: 26152519293.768
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
license: cc-by-4.0