Verfahren zur Audioverarbeitung in einem Hörgerät zur Bereitstellung einer Sprachverbesserung

Details

Veröffentlichungs-Nr.

EP4819134

Anmeldetag

31. März 2025

Veröffentlichung

7. Oktober 2026

Rechtsraum

EP

Abstract

The disclosure relates to a method of audio processing in a hearing device to provide for speech enhancement, the hearing device configured to be worn at an ear of a user, the method comprising separating one or more speech signals (SSP) from an input audio signal (SI) in a neural network (105, 145, 205, 305, 605, 635, 655), wherein an output audio signal (SO) is based on one or more of the separated speech signals (SSP). The disclosure further relates to a hearing device configured to perform the method.To provide the user with an improved natural perception of acoustic surroundings during conversations, the disclosure proposes that the method further comprises detecting, by the neural network (105, 145, 205, 305, 605, 635, 655), whether the input signal (SI) comprises an own-voice component (OVD) representative of a speech of the user, wherein, when the own-voice component (OVD) is detected, the input signal (SI) is processed so as to provide for an effective attenuation of the own-voice component (OVD) in the output signal (SO) relative to one or more of the separated speech signals (SSP) in the output signal (SO) when the own-voice component (OVD) is undetected.

Anmelder

Firma

Sonova AG

Land

🇨🇭 Schweiz