Wikibench

FLEURS

language official site →

Multilingual: no

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

Methodology

Imported from llm-stats public benchmark metadata. Modality: audio. Max score: 100. Categories: language, speech_to_text. Language: en. Verified by llm-stats: no.

Leaderboard

Showing 5 of 5 results

  1. Gemini 1.5 Flash 8B self-reported llm-stats
    86.4%
  2. Gemini 1.5 Flash self-reported llm-stats
    9.6%
  3. Gemini 1.5 Pro self-reported llm-stats
    6.7%
  4. Gemini 1.0 Pro self-reported llm-stats
    6.4%
  5. Qwen2.5-Omni-7B self-reported llm-stats
    4.1%