On Addressing Practical Challenges for RNN-Transducer

Abstract

In this paper, several works are proposed to address practical challenges for deploying RNN Transducer (RNN-T) based speech recognition system. These challenges are adapting a well-trained RNN-T model to a new domain without collecting the audio data, obtaining time stamps and confidence scores at word level. The first challenge is solved with a splicing data method which concatenates the speech segments extracted from the source domain data. To get the time stamp, a phone prediction branch is added to the RNN-T model by sharing the encoder for the purpose of force alignment. Finally, we obtain word-level confidence scores by utilizing several types of features calculated during decoding and from confusion network. Evaluated with Microsoft production data, the splicing data adaptation method improves the baseline and adaptation with the text to speech method by 58.03% and 15.25% relative word error rate reduction, respectively. The proposed time stamping method can get less than 50ms word timing difference from the ground truth alignment on average while maintaining the recognition accuracy of the RNN-T model. We also obtain high confidence annotation performance with limited computation cost.

arXiv comment5 pages

References in corpus (8)

• Sequence Transduction with Recurrent Neural Networks
• Finding consensus in speech recognition: word error minimization and other applications of confusion networks
• Adaptation Algorithms for Neural Network-Based Speech Recognition: An Overview
• Transformer-Transducer: End-to-End Speech Recognition with Self-Attention
• Modeling Multi-speaker Latent Space to Improve Neural TTS: Quick Enrolling New Speaker and Enhancing Premium Voice
• Transformer Transducer: A Streamable Speech Recognition Model with Transformer Encoders and RNN-T Loss
• Improving Performance of End-to-End ASR on Numeric Sequences
• Domain Adaptation via Teacher-Student Learning for End-to-End Speech Recognition