Streaming Sequence Transduction through Dynamic Compression
CoRR(2024)
摘要
We introduce STAR (Stream Transduction with Anchor Representations), a novel
Transformer-based model designed for efficient sequence-to-sequence
transduction over streams. STAR dynamically segments input streams to create
compressed anchor representations, achieving nearly lossless compression (12x)
in Automatic Speech Recognition (ASR) and outperforming existing methods.
Moreover, STAR demonstrates superior segmentation and latency-quality
trade-offs in simultaneous speech-to-text tasks, optimizing latency, memory
footprint, and quality.
更多查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要