arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2011.13834eess.AS
收藏

基于变换器的解码端自适应计算步长的在线语音识别

Transformer-based Online Speech Recognition with Decoder-end Adaptive Computation Steps

Mohan Li, Catalin Zorila, Rama Doddipatla

摘要:

基于转换器的端到端(E2 E)自动语音识别(ASR)系统最近获得了广泛的普及,并且在许多ASR任务上被示出优于基于递归结构的E2 E模型。然而,与其他E2 E模型一样,Transformer ASR也需要完整的输入序列来计算编码器和解码器上的注意力,这导致延迟增加,并对在线ASR构成挑战。本文提出了解码端自适应计算步骤(DACS)算法,以解决延迟的问题,并促进在线ASR。该算法通过在从编码器状态获取的置信度达到一定阈值后触发输出来流式解码Transformer ASR。不同于其他单调注意机制,风险访问整个编码器状态的每个输出步骤,本文介绍了一个最大的前瞻步骤到DACS算法,以防止到达语音结束太快。在我们的系统中采用了一个Chunkwise编码器来处理实时语音输入。建议的在线Transformer ASR系统已经在华尔街日报(WSJ)和AIShell-1数据集上进行了评估,分别产生5.5%的字错误率(WER)和7.1%的字符错误率(CER),与离线系统相比,性能只有轻微的下降。

英文摘要:

Transformer-based end-to-end (E2E) automatic speech recognition (ASR) systems have recently gained wide popularity, and are shown to outperform E2E models based on recurrent structures on a number of ASR tasks. However, like other E2E models, Transformer ASR also requires the full input sequence for calculating the attentions on both encoder and decoder, leading to increased latency and posing a challenge for online ASR. The paper proposes Decoder-end Adaptive Computation Steps (DACS) algorithm to address the issue of latency and facilitate online ASR. The proposed algorithm streams the decoding of Transformer ASR by triggering an output after the confidence acquired from the encoder states reaches a certain threshold. Unlike other monotonic attention mechanisms that risk visiting the entire encoder states for each output step, the paper introduces a maximum look-ahead step into the DACS algorithm to prevent from reaching the end of speech too fast. A Chunkwise encoder is adopted in our system to handle real-time speech inputs. The proposed online Transformer ASR system has been evaluated on Wall Street Journal (WSJ) and AIShell-1 datasets, yielding 5.5% word error rate (WER) and 7.1% character error rate (CER) respectively, with only a minor decay in performance when compared to the offline systems.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑