metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: wav_filesize
dtype: int64
- name: text
dtype: string
- name: transcript_wav2vec
dtype: string
- name: levenshtein
dtype: float64
- name: duration
dtype: float64
- name: num_words
dtype: int64
- name: speaker_id
dtype: int64
splits:
- name: train
num_bytes: 9353699242.403433
num_examples: 21519
- name: dev
num_bytes: 342735978.4656325
num_examples: 747
- name: test
num_bytes: 343721110.3959607
num_examples: 752
download_size: 9653335697
dataset_size: 10040156331.265026
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: dev
path: data/dev-*
- split: test
path: data/test-*
This is a 20% subset of the cml tts dataset (french). Audios of less than one second have been excluded.
Link to the original dataset: ylacombe/cml-tts