AI Paper Atlas

Research map / Video / Video-language understanding

HumanOmni-Speaker: Identifying Who said What and When

Detao Bai, Xihan Wei, Zhiheng Ma

ECCV 2026 · Video › Video-language understanding · approach: Video LLMs with fine-grained understanding · 4 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add Visual Delta Encoder (25 fps) (ablation)Qwen2.5-Omni-SFT · Speaker Localization (easy)miss rate (%): 3.0 -> 0.8Improved
add Visual Delta Encoder (ablation)Qwen2.5-Omni-SFT · Speaker Identification (hard)error rate (%): 33.2 -> 21.1Improved
add Visual Delta Encoder (ablation)Qwen2.5-Omni-SFT · VR-SDR (hard)SA-WER (%): 52.1 -> 47.1Improved
add Visual Delta Encoder (ablation)Qwen2.5-Omni-SFT · VR-SDR (hard)IER (%): 31.5 -> 28.5Improved
use Visual Delta Encoder alone without Visual Base Encoder (ablation)HumanOmni-Speaker · Speaker Localization (easy)miss rate (%): 3.0 -> 1.2Improved
add Visual Base Encoder to Delta Encoder (ablation)HumanOmni-Speaker · Speaker Localization (easy)miss rate (%): 1.2 -> 0.8Improved
end-to-end model instead of Whisper-diarization + Qwen3-VL pipeline (ablation)HumanOmni-Speaker · VR-SDR (hard)SA-WER (%): 64.2 -> 47.1Improved
feed full frames instead of face crops (ablation)HumanOmni-Speaker · Speaker Identification (hard)error rate (%): 38 -> 21Improved
add visual stream to audioHumanOmni-Speaker · LRS3WER (%): 3.63 -> 0.76Improved
add visual stream to audioHumanOmni-Speaker · LRS2WER (%): 3.47 -> 1.36Improved
add Delta Encoder trainingHumanOmni-Speaker · LibriSpeech test-cleanWER (%): 2.2 -> 1.88No clear effect
reduce structured tokens below 6 (ablation)Visual Delta Encoder · Speaker LocalizationaccuracyHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy