Bi-APC: Bidirectional Autoregressive Predictive Coding for Unsupervised Pre-Training and its Application to Children’s ASR
Bi-APC: Bidirectional Autoregressive Predictive Coding for Unsupervised Pre-Training and its Application to Children’s ASR
复制标题
DOI：
10.1109/icassp39728.2021.9414970
复制
发表时间：
2021-02
期刊：
影响因子：
--
通讯作者：
Ruchao Fan;Amber Afshan;A. Alwan
中科院分区：
文献类型：
--
作者：
Ruchao Fan;Amber Afshan;A. Alwan
提出了一种双向无监督模型预训练（UPT）方法，并将其应用于儿童自动语音识别（ASR）。改善儿童ASR的一个障碍是儿童语音数据库的稀缺。缓解这个问题的一种常见方法是使用成人语音数据进行模型预训练。预训练可以使用监督（SPT）或无监督方法完成，这取决于注释的可用性。通常情况下，SPT表现更好。在本文中，我们专注于UPT来解决预训练数据未标记的情况。自回归预测编码（APC）是一种UPT方法，仅从一个方向预测帧，将其用于单向预训练。然而，传统的双向UPT方法仅预测帧的一小部分。为了将APC的优点扩展到双向预训练，提出了Bi-APC。然后，我们使用自适应技术将从成人语音（使用Librisepeech语料库）学到的知识转移到儿童语音（OGI Kids语料库）。研究了基于LSTM的混杂系统。对于uni-LSTM结构，APC在基线上获得了与SPT类似的WER改进。然而，当应用于BLSTM时，APC不如SPT具有竞争力，但我们提出的Bi-APC与SPT相比具有相当的改进。
We present a bidirectional unsupervised model pre-training (UPT) method and apply it to children’s automatic speech recognition (ASR). An obstacle to improving child ASR is the scarcity of child speech databases. A common approach to alleviate this problem is model pre-training using data from adult speech. Pre-training can be done using supervised (SPT) or unsupervised methods, depending on the availability of annotations. Typically, SPT performs better. In this paper, we focus on UPT to address the situations when pre-training data are unlabeled. Autoregressive predictive coding (APC), a UPT method, predicts frames from only one direction, limiting its use to uni-directional pre-training. Conventional bidirectional UPT methods, however, predict only a small portion of frames. To extend the benefits of APC to bi-directional pre-training, Bi-APC is proposed. We then use adaptation techniques to transfer knowledge learned from adult speech (using the Librispeech corpus) to child speech (OGI Kids corpus). LSTM-based hybrid systems are investigated. For the uni-LSTM structure, APC obtains similar WER improvements to SPT over the baseline. When applied to BLSTM, however, APC is not as competitive as SPT, but our proposed Bi-APC has comparable improvements to SPT.