Temporal-Linguistic Adaptive Streaming for Continuous Sign Language Translation

Arshia Kermani, Habib Irani, Deautaun Ross, Vangelis Metsis

Keywords: Sign language translation, Simultaneous translation, Temporal pause detection, Adaptive fusion

Abstract: Real-time sign language translation must generate text incrementally as signs arrive, yet existing streaming policies treat glosses as a flat token sequence and discard the temporal rhythm of signing. Inter-gloss pauses reliably mark sentence boundaries in continuous discourse, but policies such as Wait-k cause arbitrary cross-boundary fragmentation. We propose Temporal-Linguistic Adaptive Streaming (TLAS), which fuses a Temporal Pause Detector (TPD, tracking inter-gloss interval statistics via an exponential moving average) and a Linguistic Readiness Estimator (LRE, a trained neural head on a frozen T5 encoder) through an Adaptive Fusion Gate (AFG). A proactive timeout fires before the next gloss arrives when the inter-gloss gap exceeds a threshold, producing clean sentence segmentation without oracle boundary information. We also contribute a synthetic discourse dataset of 1,400 ASL discourse groups with LLM-generated per-gloss timestamps and introduce a continuous-stream evaluation paradigm requiring autonomous boundary detection from an unbroken gloss stream. Under such conditions, TLAS significantly outperforms current heuristic baselines, such as Wait-k, and methods relying solely on linguistic content.

Submission Number: 37