---
dataset_info:
  features:
  - name: speaker_id
    dtype: int64
  - name: id
    dtype: string
  - name: transcript
    dtype: string
  - name: array
    list: float32
  - name: sampling_rate
    dtype: int64
  - name: input_text
    dtype: string
  - name: output_audio
    dtype: string
  - name: tokenized_inputs
    list: int64
  - name: tokenized_outputs
    list: int64
  splits:
  - name: validation
    num_bytes: 1392965333
    num_examples: 2703
  download_size: 650566957
  dataset_size: 1392965333
configs:
- config_name: default
  data_files:
  - split: validation
    path: data/validation-*
---