Benchmark

FLEURS

Few-shot Learning Evaluation of Universal Representations of Speech - a parallel speech dataset in 102 languages built on FLoRes-101 with approximately 12…

Modality
audio
Categories
speech_to_text, language
Openness
unknown
Source
llm_stats
Reported scores
6

Reported scores

llm_stats

ModelOrganizationReported valueReportedEvidence
Gemini 1.0 ProGoogle0.9362024-02-15evidence
Gemini 1.5 FlashGoogle0.9042024-05-01evidence
Gemini 1.5 Flash 8BGoogle0.8642024-03-15evidence
Gemini 1.5 ProGoogle0.9332024-05-01evidence
Gemma 4 12BGoogle0.9312026-05-23evidence
Qwen2.5-Omni-7BQwen0.9592025-03-27evidence

Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.