Optimizing inference performance for conformer

US12190869B2 · kind B2 · utility

Assignee

• Google LLC · US

Inventors

• Tara N. Sainath · Jersey City, US
• Rami Botros · Mountain View, US
• Anmol Gulati · New Delhi, IN
• Krzysztof Marcin Choromanski · Mountain View, US
• Ruoming Pang · New York, US
• Trevor Strohman · Sunnyvale, US
• Weiran Wang · Arlington, US
• Jiahui Yu · Champaign, US

Key dates

• Filing date: Sep 29, 2022
• Grant date: Jan 7, 2025
• Expiry date: Jun 15, 2043

Classification

• Technology area (CPC G): Physics
• CPC primary: G10L2015/223
• WIPO field: Computer technology
• WIPO sector: Electrical engineering

Abstract

A computer-implemented method includes receiving a sequence of acoustic frames as input to an automatic speech recognition (ASR) model. Here, the ASR model includes a causal encoder and a decoder. The method also includes generating, by the causal encoder, a first higher order feature representation for a corresponding acoustic frame in the sequence of acoustic frames. The method also includes generating, by the decoder, a first probability distribution over possible speech recognition hypotheses. Here, the causal encoder includes a stack of causal encoder layers each including a Recurrent Neural Network (RNN) Attention-Performer module that applies linear attention.