Multi-Stream End-to-End Speech Recognition

Abstract

Attention-based methods and Connectionist Temporal Classification (CTC) network have been promising research directions for end-to-end (E2E) Automatic Speech Recognition (ASR). The joint CTC/Attention model has achieved great success by utilizing both architectures during multi-task training and joint decoding. In this work, we present a multi-stream framework based on joint CTC/Attention E2E ASR with parallel streams represented by separate encoders aiming to capture diverse information. On top of the regular attention networks, the Hierarchical Attention Network (HAN) is introduced to steer the decoder toward the most informative encoders. A separate CTC network is assigned to each stream to force monotonic alignments. Two representative framework have been proposed and discussed, which are Multi-Encoder Multi-Resolution (MEM-Res) framework and Multi-Encoder Multi-Array (MEM-Array) framework, respectively. In MEM-Res framework, two heterogeneous encoders with different architectures, temporal resolutions and separate CTC networks work in parallel to extract complimentary information from same acoustics. Experiments are conducted on Wall Street Journal (WSJ) and CHiME-4, resulting in relative Word Error Rate (WER) reduction of 18.0-32.1% and the best WER of 3.6% in the WSJ eval92 test set. The MEM-Array framework aims at improving the far-field ASR robustness using multiple microphone arrays which are activated by separate encoders. Compared with the best single-array results, the proposed framework has achieved relative WER reduction of 3.7% and 9.7% in AMI and DIRHA multi-array corpora, respectively, which also outperforms conventional fusion strategies.

arXiv commentsubmitted to IEEE TASLP (In review). arXiv admin note: substantial text overlap with arXiv:1811.04897, arXiv:1811.04903

References in corpus (10)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Attention-Based Models for Speech Recognition
• Sequence Transduction with Recurrent Neural Networks
• Effective Approaches to Attention-based Neural Machine Translation
• Online and Linear-Time Attention by Enforcing Monotonic Alignments
• Residual Convolutional CTC Networks for Automatic Speech Recognition
• Towards better decoding and language model integration in sequence to sequence models
• Local Monotonic Attention Mechanism for End-to-End Speech and Language Processing
• Improved Regularization Techniques for End-to-End Speech Recognition
• Stream Attention for far-field multi-microphone ASR

Cited by in corpus (3)

• Multi-QuartzNet: Multi-Resolution Convolution for Speech Recognition with Multi-Layer Feature Fusion
• A practical two-stage training strategy for multi-stream end-to-end speech recognition
• Generalized Operating Procedure for Deep Learning: an Unconstrained Optimal Design Perspective