FLEURS

Few-shot Learning Evaluation of Universal Representations of Speech - a parallel speech dataset in 102 languages built on FLoRes-101 with approximately 12 hours of speech supervision per language for tasks including ASR, speech language identification, translation and retrieval

5rows
scoreprimary metric
2026-05-06sampled

Metadata

Metrics

Score, Normalized Score

Latest Results

Rank Subject Score Model Match Provenance Sampled
1 Gemini 1.5 Flash 8B 0.86 Self-reported 2026-05-06
2 Gemini 1.5 Flash 0.10 Self-reported 2026-05-06
3 Gemini 1.5 Pro 0.07 Self-reported 2026-05-06
4 Gemini 1.0 Pro 0.06 Imported 2026-05-06
5 Qwen2.5-Omni-7B 0.04 Self-reported 2026-05-06