公开(公告)号：US20250036940A1

公开(公告)日：2025-01-30

申请号：US18637161

申请日：2024-04-16

Applicant: Apple Inc.

Inventor： Ali MOIN ,  Ran LIU ,  Ellen L. ZIPPI ,  Mohammad Hadi POUR ANSARI ,  Christopher M. SANDINO ,  Erdrin AZEMI

IPC: G06N3/08

Abstract: The subject technology provides frequency-aware masked autoencoders for multimodal pretraining on frequency-based signals. An apparatus receives input data comprising frequency-based signal information associated with one or more modalities. The apparatus transforms the input data from a time domain to a frequency domain. The apparatus generates a frequency-embedded latent representation of the input data comprising time-domain and frequency-domain information. The apparatus also generates a masked frequency-embedded latent representation by masking one or more frequency components in the frequency-embedded latent representation. The apparatus produces a trained machine learning model by training a neural network to predict one or more masked frequency components of the frequency-embedded latent representation.

公开(公告)号：US20240099627A1

公开(公告)日：2024-03-28

申请号：US18369835

申请日：2023-09-18

Applicant: Apple Inc.

Inventor： Matthias R. HOHMANN ,  Ellen L. ZIPPI ,  Kaan E. DOGRUSOZ

IPC: A61B5/22 ,  A61B5/00 ,  A61B5/389

CPC classification number: A61B5/224 ,  A61B5/389 ,  A61B5/681 ,  A61B5/7225

Abstract: Aspects of the subject technology provide improved techniques for estimating muscular force. The improved techniques may include single-channel or multiple-channel surface electromyography (EMG), such as via a measurement device worn on a wrist. A muscular force estimate may be based on one or more measurements of variation between adjacent voltage measurements and estimates of spectral properties of the voltage measurements. The resulting muscular force estimate may for a basis for improved hand gesture recognition and/or heath metrics of the user.