metadata
dataset_info:
features:
- name: transcript
struct:
- name: duration
dtype: float64
- name: language
dtype: string
- name: segments
list:
- name: chars
dtype: 'null'
- name: end
dtype: float64
- name: id
dtype: int64
- name: start
dtype: float64
- name: text
dtype: string
- name: words
list:
- name: end
dtype: float64
- name: score
dtype: float64
- name: start
dtype: float64
- name: word
dtype: string
- name: task
dtype: string
- name: text
dtype: string
- name: words
list:
- name: end
dtype: float64
- name: score
dtype: float64
- name: start
dtype: float64
- name: word
dtype: string
- name: pronunciation_analysis
struct:
- name: eff_per
dtype: float64
- name: ielts_score
dtype: int64
- name: pronunciation_score
dtype: int64
- name: r_blurred
dtype: float64
- name: r_broken
dtype: float64
- name: r_clear
dtype: float64
- name: segments
list:
- name: chars
dtype: 'null'
- name: eff_per
dtype: float64
- name: end
dtype: float64
- name: ielts_score
dtype: int64
- name: pronunciation_score
dtype: int64
- name: r_blurred
dtype: float64
- name: r_broken
dtype: float64
- name: r_clear
dtype: float64
- name: start
dtype: float64
- name: text
dtype: string
- name: words
list:
- name: end
dtype: float64
- name: phonemes
list:
- name: base_phoneme
dtype: string
- name: end
dtype: float64
- name: error_type
dtype: string
- name: grapheme
dtype: string
- name: mispronounced
dtype: bool
- name: phoneme
dtype: string
- name: pronunciation_score
dtype: float64
- name: required_attributes
list:
- name: attribute
dtype: string
- name: error_type
dtype: string
- name: present
dtype: bool
- name: sounds_like
dtype: string
- name: start
dtype: float64
- name: unexpected_attributes
list:
- name: attribute
dtype: string
- name: error_type
dtype: string
- name: present
dtype: bool
- name: variation_rule
struct:
- name: canonical_phone
dtype: string
- name: category
dtype: string
- name: realized_phone
dtype: string
- name: rule_id
dtype: string
- name: rule_name
dtype: string
- name: pronunciation_score
dtype: int64
- name: sounds_like
dtype: string
- name: start
dtype: float64
- name: word
dtype: string
- name: word_segments
list:
- name: end
dtype: float64
- name: phonemes
list:
- name: base_phoneme
dtype: string
- name: end
dtype: float64
- name: error_type
dtype: string
- name: grapheme
dtype: string
- name: mispronounced
dtype: bool
- name: phoneme
dtype: string
- name: pronunciation_score
dtype: float64
- name: required_attributes
list:
- name: attribute
dtype: string
- name: error_type
dtype: string
- name: present
dtype: bool
- name: sounds_like
dtype: string
- name: start
dtype: float64
- name: unexpected_attributes
list:
- name: attribute
dtype: string
- name: error_type
dtype: string
- name: present
dtype: bool
- name: variation_rule
struct:
- name: canonical_phone
dtype: string
- name: category
dtype: string
- name: realized_phone
dtype: string
- name: rule_id
dtype: string
- name: rule_name
dtype: string
- name: pronunciation_score
dtype: int64
- name: sounds_like
dtype: string
- name: start
dtype: float64
- name: word
dtype: string
- name: pronunciation_analysis_text
dtype: string
- name: llm_score
dtype: string
- name: llm_score__reasoning_content
dtype: string
splits:
- name: train
num_bytes: 58843665
num_examples: 500
download_size: 9798906
dataset_size: 58843665
configs:
- config_name: default
data_files:
- split: train
path: data/train-*