Convolutional recurrent neural networks for small-footprint keyword spotting

US10540961B2 · kind B2 · utility

Assignee

• BAIDU USA LLC · US

Inventors

• Sercan Omer Arik · San Francisco, US
• Markus Kliegl · Santa Clara, US
• Rewon Child · Emeryville, US
• Joel Hestness · Mountain View, US
• Andrew Gibiansky · Mountain View, US
• Christopher Fougner · London, GB
• Ryan Prenger · Oakland, US
• Adam Coates · Mountain View, US

Key dates

• Filing date: Aug 28, 2017
• Grant date: Jan 21, 2020
• Expiry date: Dec 23, 2037

Classification

• Technology area (CPC G): Physics
• CPC primary: G10L2015/088
• WIPO field: Computer technology
• WIPO sector: Electrical engineering

Abstract

Described herein are systems and methods for creating and using Convolutional Recurrent Neural Networks (CRNNs) for small-footprint keyword spotting (KWS) systems. Inspired by the large-scale state-of-the-art speech recognition systems, in embodiments, the strengths of convolutional layers to utilize the structure in the data in time and frequency domains are combined with recurrent layers to utilize context for the entire processed frame. The effect of architecture parameters were examined to determine preferred model embodiments given the performance versus model size tradeoff. Various training strategies are provided to improve performance. In embodiments, using only ˜230 k parameters and yielding acceptably low latency, a CRNN model embodiment demonstrated high accuracy and robust performance in a wide range of environments.