{"benchmark_id":"fleurs","name":"FLEURS","parent_benchmark":null,"categories":["speech_to_text","language"],"modality":"audio","multilingual":true,"max_score":1.0,"language":"en","description":"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. Scores are shown as speech recognition accuracy (1 - word error rate), so higher is better.","paper_link":"https://arxiv.org/abs/2205.12446","implementation_link":null,"verified":false,"created_at":"2026-05-07T16:53:23.225736+00:00","updated_at":"2026-07-05T18:27:48.051220+00:00","statistics":{"total_models":6,"average_score":0.9211666666666667,"min_score":0.864,"max_score":0.959,"score_stddev":0.03301767203584568,"verified_count":0,"self_reported_count":5},"child_benchmarks":[],"linked_dataset":null,"models":[{"rank":1,"model_id":"qwen2.5-omni-7b","model_name":"Qwen2.5-Omni-7B","organization_id":"qwen","organization_name":"Alibaba Cloud / Qwen Team","organization_country":"CN","score":0.959,"normalized_score":0.959,"verified":false,"self_reported":true,"self_reported_source":"https://github.com/QwenLM/Qwen2.5-Omni","analysis_method":"Speech recognition accuracy (1 - WER)","verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2025-03-27","param_count":7000000000,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null},{"rank":2,"model_id":"gemini-1.0-pro","model_name":"Gemini 1.0 Pro","organization_id":"google","organization_name":"Google","organization_country":"US","score":0.936,"normalized_score":0.936,"verified":false,"self_reported":false,"self_reported_source":"https://example.com/benchmark-image","analysis_method":"Speech recognition accuracy (1 - WER)","verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2024-02-15","param_count":null,"is_open_source":false,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null},{"rank":3,"model_id":"gemini-1.5-pro","model_name":"Gemini 1.5 Pro","organization_id":"google","organization_name":"Google","organization_country":"US","score":0.933,"normalized_score":0.933,"verified":false,"self_reported":true,"self_reported_source":"https://deepmind.google/technologies/gemini/pro/","analysis_method":"Speech recognition accuracy (1 - WER)","verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2024-05-01","param_count":null,"is_open_source":false,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null},{"rank":4,"model_id":"gemma-4-12b-it","model_name":"Gemma 4 12B","organization_id":"google","organization_name":"Google","organization_country":"US","score":0.931,"normalized_score":0.931,"verified":false,"self_reported":true,"self_reported_source":"https://huggingface.co/google/gemma-4-12B-it","analysis_method":"Speech recognition accuracy (1 - WER)","verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2026-05-23","param_count":11959730224,"is_open_source":true,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null},{"rank":5,"model_id":"gemini-1.5-flash","model_name":"Gemini 1.5 Flash","organization_id":"google","organization_name":"Google","organization_country":"US","score":0.904,"normalized_score":0.904,"verified":false,"self_reported":true,"self_reported_source":"https://deepmind.google/technologies/gemini/flash/","analysis_method":"Speech recognition accuracy (1 - WER)","verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2024-05-01","param_count":null,"is_open_source":false,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null},{"rank":6,"model_id":"gemini-1.5-flash-8b","model_name":"Gemini 1.5 Flash 8B","organization_id":"google","organization_name":"Google","organization_country":"US","score":0.864,"normalized_score":0.864,"verified":false,"self_reported":true,"self_reported_source":"https://deepmind.google/technologies/gemini/flash/","analysis_method":"Speech recognition accuracy (1 - WER)","verification_date":null,"provider_id":null,"input_cost_per_million":null,"output_cost_per_million":null,"context_window":null,"announcement_date":"2024-03-15","param_count":8000000000,"is_open_source":false,"is_new":false,"best_latency":null,"latency_provider":null,"best_throughput":null,"throughput_provider":null,"context_provider":null}]}