Unverified50% confidenceBenchmarkExact time
Marimba achieves decoder AUROC of 0.80–0.92 on a private conversation test set and 0.92–0.97 on public read speech.
1
Sources
50%
Confidence
Long-term
Relevance
11/16/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedMarimba's personal information detection achieves an entry-level F1 of 0.904 on synthetic Brazilian Portuguese speech and 1.000 on human reference transcripts.70% similarUnverifiedQwen-Audio-Jev achieves an equal error rate (EER) of 6.50% on the VoxCeleb1-O speaker verification benchmark.64% similarUnverifiedMedAstra's pure speech retriever achieves 87.6 Recall@10, the hybrid configuration achieves 90.4, and transcription-first dense retrieval achieves 81.0.63% similarUnverifiedExisting LLM-driven Omni embedding models score 95-99% on multilingual text-to-audio retrieval on the MAEB leaderboard, but only 1%-6% on unsupervised audio clustering and about 20% on speech emotion recognition (CREMA-D).61% similarUnverifiedOn both the FLEURS and Common Voice datasets, ufakzeka-kulak's WER is lower than Whisper base and Whisper small, and the paired confidence intervals are entirely below zero, so the performance advantage is statistically significant.60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1173842API
curl https://kongchang.com/api/v1/knowledge/claims/1173842MCP
get_claim(id=1173842)