Unverified50% confidenceBenchmarkExact time
Qwen-Audio-Jev achieves an equal error rate (EER) of 6.50% on the VoxCeleb1-O speaker verification benchmark.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/16/2026
First Seen
Valid until: 1/14/2027
Sources
Related Entities
Related Claims
UnverifiedThe latest version of pyannote.audio adopts an end-to-end neural network architecture (EEND), and the diarization error rate (DER) on standard evaluation sets has dropped below 10%.70% similarUnverifiedQwen-Audio-Jev achieves a CHOICE accuracy of 84.06% on MMSU.66% similarUnverifiedQ8_0（8bit 量化）的精度损失对语音识别任务几乎可以忽略66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1047457API
curl https://kongchang.com/api/v1/knowledge/claims/1047457MCP
get_claim(id=1047457)