Transformer-Transducer: End-to-End Speech Recognition with Self-Attention

Abstract

We explore options to use Transformer networks in neural transducer for end-to-end speech recognition. Transformer networks use self-attention for sequence modeling and comes with advantages in parallel computation and capturing contexts. We propose 1) using VGGNet with causal convolution to incorporate positional information and reduce frame rate for efficient inference 2) using truncated self-attention to enable streaming for Transformer and reduce computational complexity. All experiments are conducted on the public LibriSpeech corpus. The proposed Transformer-Transducer outperforms neural transducer with LSTM/BLSTM networks and achieved word error rates of 6.37 % on the test-clean set and 15.30 % on the test-other set, while remaining streamable, compact with 45.7M parameters for the entire system, and computationally efficient with complexity of O(T), where T is input sequence length.

References in corpus (2)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Sequence Transduction with Recurrent Neural Networks

Cited by in corpus (24)

• Thank you for Attention: A survey on Attention-based Artificial Neural Networks for Automatic Speech Recognition
• UniSpeech at scale: An Empirical Study of Pre-training Method on Large-Scale Speech Recognition Dataset
• A Further Study of Unsupervised Pre-training for Transformer Based Speech Recognition
• Attention-based Transducer for Online Speech Recognition
• Weak-Attention Suppression For Transformer Based Speech Recognition
• Transformer-based Arabic Dialect Identification
• Multi-view Frequency LSTM: An Efficient Frontend for Automatic Speech Recognition
• Efficient End-to-End Speech Recognition Using Performers in Conformers
• Conv-Transformer Transducer: Low Latency, Low Frame Rate, Streamable End-to-End Speech Recognition
• Adversarial Joint Training with Self-Attention Mechanism for Robust End-to-End Speech Recognition
• Learning Word-Level Confidence For Subword End-to-End ASR
• Unidirectional Memory-Self-Attention Transducer for Online Speech Recognition
• A Light-weight contextual spelling correction model for customizing transducer-based speech recognition systems
• Cross-attention conformer for context modeling in speech enhancement for ASR
• Streaming Transformer-based Acoustic Models Using Self-attention with Augmented Memory
• Flexi-Transducer: Optimizing Latency, Accuracy and Compute forMulti-Domain On-Device Scenarios
• Context-Aware Transformer Transducer for Speech Recognition
• Multi-Channel Transformer Transducer for Speech Recognition
• CarneliNet: Neural Mixture Model for Automatic Speech Recognition
• Dynamic Encoder Transducer: A Flexible Solution For Trading Off Accuracy For Latency
• Multi-mode Transformer Transducer with Stochastic Future Context
• Transformer Based Deliberation for Two-Pass Speech Recognition
• FSR: Accelerating the Inference Process of Transducer-Based Models by Applying Fast-Skip Regularization
• Multitask Learning and Joint Optimization for Transformer-RNN-Transducer Speech Recognition