Voice wake-up method and device, as well as storage medium and system

DE602022022021T2Active Publication Date: 2025-09-24HUAWEI TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
DE602022022021
Authority / Receiving Office
DE · DE
Patent Type
Patents
Current Assignee / Owner
Priority Date
2021-03-31
Filing Date
2022-03-25
Publication Date
2025-09-24
Estimated Expiration
2042-03-25

AI Technical Summary

Technical Problem

Existing voice wakeup systems in electronic devices suffer from poor performance in complex acoustic environments, particularly with background noise, multi-sound source interference, and far-field echoes, leading to low wakeup rates and high false wakeups.

Method used

A two-stage separation and wakeup solution is implemented, where a first-stage streaming process uses a neural network to separate and determine potential wakeups, followed by a second-stage offline process to confirm and reduce false wakeups, employing dpconformer network structures for both stages.

Benefits of technology

This approach significantly enhances wakeup rates while minimizing false wakeups, ensuring reliable voice interaction in challenging acoustic conditions.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader
Need to check novelty before this filing date? Find Prior Art
No text content released.