/Patent Tracker
GooglePatent applicationAI & softwareInterest 6/10

Google patents image tricks to help computers hear better

Patent drawing
Patent drawing from the official filing.

Google describes a method for improving speech recognition software by treating sound as a picture. The system creates new training data by hiding random sections of these sound pictures, forcing the computer to learn to understand audio even when parts are missing or noisy.

Why it mattersIt offers a specific technical approach to making voice assistants more reliable in real-world environments where background noise is common.

In the filing’s own words

Generally, the present disclosure is directed to systems and methods that generate augmented training data for machine-learned models via application of one or more augmentation techniques to audiographic images that visually represent audio signals. In particular, the present disclosure provides a number of novel augmentation operations which can be performed directly upon the audiographic image (e.g., as opposed to the raw audio data) to generate augmented training data that results in improved model performance. As an example, the audiographic images can be or include one or more spectrograms or filter bank sequences.

What exactly they claim (claim 1)

1. A computer-implemented method of data augmentation for training machine-learned models, the method comprising: obtaining a plurality of audiographic images that represent audio signals using pixels; for one or more of the audiographic images, generating one or more augmented images based on masking one or more blocks of consecutive pixel values along a time or frequency axis, the quantity of the one or more blocks of consecutive pixel values defined by a hyperparameter, each respective block having a respective width and a respective location; and training a machine-learned model on the augmented images to output text transcriptions by modifying model parameters based on an evaluation of an objective function.

Facts

Company
Google (Google LLC)
Official title
Augmentation of Audiographic Images for Improved Machine Learning
Publication number
US20260310968A1
Application number
19711519
Filed
Jun 17, 2026
Published
Oct 8, 2026
Inventors
7
Claims
20

Classified under

G06N 3/084 Physics & computingG06F 18/2148 Physics & computingG06N 20/00 Physics & computingG06V 10/7747 Physics & computingG06V 10/82 Physics & computingG10L 15/063 Physics & computing
Read the full filing (official PDF) ↗ Open in the Patent Tracker

More from Google in AI & software

Summaries are simplified from the filing’s own abstract and first claim. A patent filing is not a product announcement.
Patent data and drawings: U.S. Patent and Trademark Office (USPTO). Nebula is not affiliated with or endorsed by the USPTO.