Echo: A Joint-Embedding Predictive Architecture for Speaker Diarization and Speech Recognition in a Shared Latent Space 文章

ArXiv CS.AI2026-06-02NEWSen作者: Louis Mouchon
查看原文 →
关系图谱

Echo: A Joint-Embedding Predictive Architecture for Speaker Diarization and Speech Recognition in a Shared Latent Space · 相关技术

相关技术

k-NNViTAnull-target K-set predictionPITJEPA objectivek-NN probeVBxArcFaceJEPAViT encoder