arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2211.00786cs.SDcs.CLeess.AS
收藏

用于快速高效语音系统的统一端到端语音识别和端点定位

Unified End-to-End Speech Recognition and Endpointing for Fast and Efficient Speech Systems

Google, Inc.(谷歌公司)

Shaan Bijwadia, Shuo-yiin Chang, Bo Li, Tara Sainath, Chao Zhang, Yanzhang He

摘要:

自动语音识别(ASR)系统通常依赖于外部端点(EP)模型来识别语音边界。在这项工作中,我们提出了一种方法,联合训练的ASR和EP任务在一个单一的端到端(E2E)多任务模型,提高EP质量,可选地利用信息从ASR音频编码器。我们引入了一个“开关”连接,它训练EP直接消费音频帧或来自ASR模型的低级潜在表示。这导致单个E2E模型可以在推理期间用于以低成本执行帧过滤,并且还基于正在进行的ASR计算进行高质量的查询结束(EOQ)预测。我们目前的语音搜索测试集上的结果表明,与单独的单任务模型相比,这种方法减少了120毫秒(减少30.8%)的中位端点延迟,和第90百分位延迟170毫秒(减少23.0%),而不会回归字出错率。对于连续识别,WER提高了10.6%(相对)。

英文摘要:

Automatic speech recognition (ASR) systems typically rely on an external endpointer (EP) model to identify speech boundaries. In this work, we propose a method to jointly train the ASR and EP tasks in a single end-to-end (E2E) multitask model, improving EP quality by optionally leveraging information from the ASR audio encoder. We introduce a "switch" connection, which trains the EP to consume either the audio frames directly or low-level latent representations from the ASR model. This results in a single E2E model that can be used during inference to perform frame filtering at low cost, and also make high quality end-of-query (EOQ) predictions based on ongoing ASR computation. We present results on a voice search test set showing that, compared to separate single-task models, this approach reduces median endpoint latency by 120 ms (30.8% reduction), and 90th percentile latency by 170 ms (23.0% reduction), without regressing word error rate. For continuous recognition, WER improves by 10.6% (relative).

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。

↑