Phoneme Based Neural Transducer for Large Vocabulary Speech Recognition
arXiv:2010.16368
Abstract
To join the advantages of classical and end-to-end approaches for speech recognition, we present a simple, novel and competitive approach for phoneme-based neural transducer modeling. Different alignment label topologies are compared and word-end-based phoneme label augmentation is proposed to improve performance. Utilizing the local dependency of phonemes, we adopt a simplified neural network structure and a straightforward integration with the external word-level language model to preserve the consistency of seq-to-seq modeling. We also present a simple, stable and efficient training procedure using frame-wise cross-entropy loss. A phonetic context size of one is shown to be sufficient for the best performance. A simplified scheduled sampling approach is applied for further improvement and different decoding approaches are briefly compared. The overall performance of our best model is comparable to state-of-the-art (SOTA) results for the TED-LIUM Release 2 and Switchboard corpora.
arXiv commentaccepted at ICASSP2021
References in corpus (11)
- SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition
- Scheduled Sampling for Sequence Prediction with Recurrent Neural Networks
- Sequence Transduction with Recurrent Neural Networks
- A Comparative Study on Transformer vs RNN in Speech Applications
- On Using Monolingual Corpora in Neural Machine Translation
- A Comprehensive Study of Deep Bidirectional LSTM RNNs for Acoustic Modeling in Speech Recognition
- The CAPIO 2017 Conversational Speech Recognition System
- A New Training Pipeline for an Improved Neural Transducer
- On the Choice of Modeling Unit for Sequence-to-Sequence Speech Recognition
- An investigation of phone-based subword units for end-to-end speech recognition
- Context-Dependent Acoustic Modeling without Explicit Phone Clustering