Supervised and Unsupervised Training with Contrastive Loss Over Sequences
AssigneeGoogle Inc.
InventorsAndrew Rosenberg, Bhuvana Ramabhadran, Zhehuai Chen, Yuan Wang, Yu Zhang, Jesse Emond
Life of this patent
Estimating the term…
Loading patent…
AssigneeGoogle Inc.
InventorsAndrew Rosenberg, Bhuvana Ramabhadran, Zhehuai Chen, Yuan Wang, Yu Zhang, Jesse Emond
Estimating the term…
A method includes receiving audio data corresponding to an utterance and generating a pair of positive audio data examples. Here, each positive audio data example includes a respective augmented copy of the received audio data. For each respective positive audio data example, the method includes generating a respective sequence of encoder outputs and projecting the respective sequence of encoder outputs for the positive data example into a contrastive loss space. The method also includes determining a L2 distance between each corresponding encoder output in the projected sequences of encoder outputs for the positive audio data examples and determining a per-utterance consistency loss by averaging the L2 distances. The method also includes generating corresponding speech recognition results for each respective positive audio data example. The method also includes updating parameters of the speech recognition model based on a respective supervised loss term and the per-utterance consistency loss.
A computer-implemented method executing on data processing hardware that causes the data processing hardware to perform operations comprising:
A system comprising:
Dependencies are inferred from the claim text.