New paper proposes recurrence-based approach to ultra-efficient on-device speech recognition
An arXiv paper, 'Depth through Recurrence: Towards Ultra-Efficient On-Device ASR,' proposes using recurrence to achieve deep-model accuracy in speech recognition while cutting memory and compute costs for on-device deployment.
Read the full story at arXiv cs.LG ↗
Timeline · 1 report
- 2026-10-30 04:00 · arXiv cs.LG
Depth through Recurrence: Towards Ultra-Efficient On-Device ASR