globe_v2
ASR-ready version of MushanW/GLOBE_V2 (license: CC0-1.0).
Common Voice 14 English, enhanced. ~69% US English; filter on accent for non-US/UK speakers. Overlaps common_accent (same Common Voice source).
Format: parquet, 500-row row groups (Cohere training format). Columns: audio (bytes, 16 kHz mono FLAC), id, text,
duration, accent, speaker_id, gender, source, lang, reference_text, meta (JSON string).
Clips: 0.5–30 s, non-empty text. No judge/quality filtering applied.
Splits: test: 5,042 clips / 4.9 h, train: 533,405 clips / 506.3 h, validation: 3,779 clips / 3.7 h
Top accent labels: United States English 343,534, England English 88,111, Canadian English 38,888, Australian English 18,164, Northern Irish 11,515, India and South Asia (India, Pakistan, Sri Lanka) 10,633, German English,Non native speaker 8,081, Filipino 3,352, Scottish English 2,762, England English,New Zealand English 2,715, New Zealand English 2,685, Irish English 2,658, Southern African (South Africa, Zimbabwe, Namibia) 1,183, Hong Kong English 1,014, United States English,southern United States,New Orleans dialect 554
- Downloads last month
- 20