{"ID":2921051,"CreatedAt":"2026-06-02T02:42:49.606572591Z","UpdatedAt":"2026-06-04T07:41:34.29888543Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2606.01909","arxiv_id":"2606.01909","title":"Echo: A Joint-Embedding Predictive Architecture for Speaker Diarization and Speech Recognition in a Shared Latent Space","abstract":"We present Echo, a proof-of-concept audio system built around a single 25 M-parameter ViT encoder. The encoder is pretrained with a JEPA objective and then specialised by stages to carry speaker identity, phonetic content, and dynamic source routing in the same 512-dimensional latent space, with no per-task fine-tuning at deployment. Light heads handle diarization (ArcFace + VBx) and dynamic source separation (null-target K-set prediction). On synthetic VoxCeleb2 mixtures with unknown K, the canonical stack reaches 15.00% blind DER, 97.80% PIT separation accuracy with +9.52 dB latent SI-SDR, and a +53.50-point speaker/content factorisation gap on a held-out k-NN probe. The point of Echo is not a new SOTA on any single task but the joint coexistence of three tasks on one encoder at this footprint. We document the design stage by stage, report the dead-ends, and identify the structural wall on end-to-end ASR through the VQ bottleneck that still bounds the PoC.","short_abstract":"We present Echo, a proof-of-concept audio system built around a single 25 M-parameter ViT encoder. The encoder is pretrained with a JEPA objective and then specialised by stages to carry speaker identity, phonetic content, and dynamic source routing in the same 512-dimensional latent space, with no per-task fine-tuning...","url_abs":"https://arxiv.org/abs/2606.01909","url_pdf":"https://arxiv.org/pdf/2606.01909v1","authors":"[\"Louis Mouchon\"]","published":"2026-06-01T08:46:11Z","proceeding":"cs.SD","tasks":"[\"cs.SD\",\"cs.AI\",\"eess.AS\"]","methods":"[]","has_code":false}