Work

Hybrid Autoregressive Transducer (HAT)

Year: 2020

Type: conference-paper

Abstract: This paper proposes and evaluates the hybrid autoregressive transducer (HAT) model, a time-synchronous encoder-decoder model that preserves the modularity of conventional automatic speech recognition systems. The HAT model provides a way to measure the quality of the internal language model that can be used to decide whether inference with an external language model is beneficial or not. We evaluate our proposed model on a large-scale voice search task. Our experiments show significant improvements in WER compared to the state-of-the-art approaches1.

Source: IEEE International Conference on Acoustics Speech and Signal Processing

DOI: 10.1109/icassp40776.2020.9053600

Authors: Ehsan Variani, David Rybach, Cyril Allauzen, Michael Riley

Institutions: Google (United States), AT&T (United States)

Language: English

Topic: Speech Recognition and Synthesis

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

OpenAlex ID: W3011956499

FWCI: 9.828

Cites: 44

Cited by: 132

Related to: 20