TrimTail: Low-Latency Streaming ASR with Simple but Effective Spectrogram-Level Length Penalty
| Autor: | Xingchen Song, Di Wu, Zhiyong Wu, Binbin Zhang, Yuekai Zhang, Zhendong Peng, Wenpeng Li, Fuping Pan, Changbao Zhu |
|---|---|
| Rok vydání: | 2022 |
| Předmět: |
FOS: Computer and information sciences
Sound (cs.SD) Computer Science - Computation and Language Audio and Speech Processing (eess.AS) I.2.7 FOS: Electrical engineering electronic engineering information engineering Computation and Language (cs.CL) Computer Science - Sound Electrical Engineering and Systems Science - Audio and Speech Processing |
| DOI: | 10.48550/arxiv.2211.00522 |
| Popis: | In this paper, we present TrimTail, a simple but effective emission regularization method to improve the latency of streaming ASR models. The core idea of TrimTail is to apply length penalty (i.e., by trimming trailing frames, see Fig. 1-(b)) directly on the spectrogram of input utterances, which does not require any alignment. We demonstrate that TrimTail is computationally cheap and can be applied online and optimized with any training loss or any model architecture on any dataset without any extra effort by applying it on various end-to-end streaming ASR networks either trained with CTC loss [1] or Transducer loss [2]. We achieve 100 $\sim$ 200ms latency reduction with equal or even better accuracy on both Aishell-1 and Librispeech. Moreover, by using TrimTail, we can achieve a 400ms algorithmic improvement of User Sensitive Delay (USD) with an accuracy loss of less than 0.2. Comment: submitted to ICASSP 2023 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|