Streaming Sequence Transduction through Dynamic Compression
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913849102303232 |
|---|---|
| author | Tan, Weiting Chen, Yunmo Chen, Tongfei Qin, Guanghui Xu, Haoran Zhang, Heidi C. Van Durme, Benjamin Koehn, Philipp |
| author_facet | Tan, Weiting Chen, Yunmo Chen, Tongfei Qin, Guanghui Xu, Haoran Zhang, Heidi C. Van Durme, Benjamin Koehn, Philipp |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_01172 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Streaming Sequence Transduction through Dynamic Compression Tan, Weiting Chen, Yunmo Chen, Tongfei Qin, Guanghui Xu, Haoran Zhang, Heidi C. Van Durme, Benjamin Koehn, Philipp Computation and Language Sound Audio and Speech Processing 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. |
| title | Streaming Sequence Transduction through Dynamic Compression |
| topic | Computation and Language Sound Audio and Speech Processing |
| url | https://arxiv.org/abs/2402.01172 |