提出Spiralformer模型，降低语音识别编码延迟。
Spiralformer: Low Latency Encoder for Streaming Speech Recognition with Circular Layer Skipping and Early Exiting
- 通过循环跳层与提前退出机制，实现小块高频输出
- 在Librispeech上平均发音延迟减少21.6%
- 适合对实时性要求高的流式语音识别场景
针对流式语音识别中的编码延迟问题，现有基于Transformer的编码器多采用块处理方式。尽管已有研究关注解码器的发射延迟优化，但对块处理中编码延迟的改进仍较少。本文提出Spiralformer，通过周期性跳过层计算并螺旋式移动已计算层，在保持相似计算成本和词错误率的前提下，实现更频繁的小块输出。实验表明，在Librispeech数据集上，该方法使平均令牌发射延迟降低21.6%；在CSJ数据集上降低7.0%，显著提升了实时性。
原文摘要 · Abstract (English)
For streaming speech recognition, a Transformer-based encoder has been widely used with block processing. Although many studies addressed improving emission latency of transducers, little work has been explored for improving encoding latency of the block processing. We seek to reduce latency by frequently emitting a chunk with a small shift rather than scarce large-chunk emissions, resulting in higher computational costs. To efficiently compute with the small chunk shift, we propose a new encoder, Spiralformer, tailored for block processing by combining layer dropping and early exiting. We skip layer computation in a cyclic manner and shift the computed layer in each block spirally, which completes computation for all the layers over the block processing. Experimentally, we observed that our method achieved 21.6% reduction in the averaged token emission delay in Librispeech, and 7.0% in CSJ, compared with the baseline with similar computational cost and word error rates.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。