---
dataset_info:
  features:
  - name: text
    dtype: string
  - name: speech_tokens
    sequence: int64
  - name: style_emb
    sequence: float32
  - name: speaker_id
    dtype: string
  splits:
  - name: train
    num_bytes: 36052288
    num_examples: 30000
  download_size: 30760287
  dataset_size: 36052288
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---