Streaming End-to-end Speech Recognition For Mobile Devices

Abstract

End-to-end (E2E) models, which directly predict output character sequences given input speech, are good candidates for on-device speech recognition. E2E models, however, present numerous challenges: In order to be truly useful, such models must decode speech utterances in a streaming fashion, in real time; they must be robust to the long tail of use cases; they must be able to leverage user-specific context (e.g., contact lists); and above all, they must be extremely accurate. In this work, we describe our efforts at building an E2E speech recognizer using a recurrent neural network transducer. In experimental evaluations, we find that the proposed approach can outperform a conventional CTC-based model in terms of both latency and accuracy in a number of evaluation categories.

References in corpus (3)

• Sequence Transduction with Recurrent Neural Networks
• EESEN: End-to-End Speech Recognition using Deep RNN Models and WFST-based Decoding
• Personalized Speech recognition on mobile devices

Cited by in corpus (6)

• An Electro-Photonic System for Accelerating Deep Neural Networks
• A Simplified Fully Quantized Transformer for End-to-end Speech Recognition
• Efficient End-to-End Speech Recognition Using Performers in Conformers
• Exploiting Large-scale Teacher-Student Training for On-device Acoustic Models
• End-to-End Automatic Speech Recognition Integrated With CTC-Based Voice Activity Detection
• Hybrid phonetic-neural model for correction in speech recognition systems