---
dataset_info:
  features:
  - name: context
    dtype: audio
  - name: instruction
    dtype: string
  - name: speech_instruction
    dtype: string
  - name: answer
    dtype: string
  - name: sound_tokens
    dtype: string
  splits:
  - name: test
    num_bytes: 14018734.0
    num_examples: 100
  download_size: 12281950
  dataset_size: 14018734.0
configs:
- config_name: default
  data_files:
  - split: test
    path: data/test-*
---