---
dataset_info:
  features:
  - name: context
    dtype:
      audio:
        sampling_rate: 192000
  - name: instruction
    dtype: string
  - name: speech_instruction
    dtype: string
  - name: answer
    dtype: string
  splits:
  - name: test
    num_bytes: 228634291.0
    num_examples: 100
  download_size: 197814930
  dataset_size: 228634291.0
configs:
- config_name: default
  data_files:
  - split: test
    path: data/test-*
---