The patent badge is an abbreviated version of the USPTO patent document. The patent badge does contain a link to the full patent document.
The patent badge is an abbreviated version of the USPTO patent document. The patent badge covers the following: Patent number, Date patent was issued, Date patent was filed, Title of the patent, Applicant, Inventor, Assignee, Attorney firm, Primary examiner, Assistant examiner, CPCs, and Abstract. The patent badge does contain a link to the full patent document (in Adobe Acrobat format, aka pdf). To download or print any patent click here.
Patent No.:
Date of Patent:
May. 19, 2020
Filed:
Jan. 30, 2018
Baidu Usa, Llc, Sunnyvale, CA (US);
Eric Battenberg, Sunnyvale, CA (US);
Rewon Child, San Francisco, CA (US);
Adam Coates, Mountain View, CA (US);
Christopher Fougner, Palo Alto, CA (US);
Yashesh Gaur, Santa Clara, CA (US);
Jiaji Huang, San Jose, CA (US);
Heewoo Jun, Sunnyvale, CA (US);
Ajay Kannan, Mountain View, CA (US);
Markus Kliegl, Santa Clara, CA (US);
Atul Kumar, Campbell, CA (US);
Hairong Liu, San Jose, CA (US);
Vinay Rao, Mountain View, CA (US);
Sanjeev Satheesh, Sunnyvale, CA (US);
David Seetapun, Berkeley, CA (US);
Anuroop Sriram, Sunnyvale, CA (US);
Zhenyao Zhu, Sunnyvale, CA (US);
Baidu USA LLC, Sunnyvale, CA (US);
Abstract
Described herein are systems and methods to identify and address sources of bias in an end-to-end speech model. In one or more embodiments, the end-to-end model may be a recurrent neural network with two 2D-convolutional input layers, followed by multiple bidirectional recurrent layers and one fully connected layer before a softmax layer. In one or more embodiments, the network is trained end-to-end using the CTC loss function to directly predict sequences of characters from log spectrograms of audio. With optimized recurrent layers and training together with alignment information, some unwanted bias induced by using purely forward only recurrences may be removed in a deployed model.