Unverified50% confidenceBenchmarkExact time
On the LibriSpeech benchmark, Oído's word error rate (WER) is 3.7 (clean) and 8.2 (other), while Whisper tiny.en on a laptop achieves 6.3 and 15.9.
1
Sources
50%
Confidence
Long-term
Relevance
9/30/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedWhisper has been iterated to the large-v3 version (November 2023), with an average word error rate (WER) on the Fleurs benchmark reduced to approximately below 10%.75% similarUnverifiedOn the LibriSpeech test-clean benchmark, Whisper large-v3's WER has dropped to around 2%, approaching or in some cases surpassing human annotators (approximately 4–5%).69% similarUnverifiedLing 3.0 Tiny 每处理一个token的算术运算量约为8B稠密模型的六分之一62% similarVerifiedWhisper is available in multiple sizes ranging from tiny (39M parameters) to large-v3 (1.5B parameters).61% similarUnverifiedMistral Small 4 scored only 0.699 on unit conversion inputs and produced 265 errors that differed from the correct label by exactly a power of ten.59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/962422API
curl https://kongchang.com/api/v1/knowledge/claims/962422MCP
get_claim(id=962422)