Echounterdrücker zur Verwendung in der Bidirektionalen Sprachkommunikation

Details

Veröffentlichungs-Nr.

EP4801022

Anmeldetag

16. Januar 2026

Veröffentlichung

2. September 2026

Rechtsraum

EP

IPC

H04M9/08

Abstract

An echo suppressor (10) for use in two-way voice communication through a communication network (4) in a scenario wherein near-end speech (SN) from a local user (1) is picked up by a microphone (8) for transmission to a remote participant (6) and far-end speech (RX) received from the remote participant (6) is reproduced for the local user (1) by a loudspeaker (7).The echo suppressor comprises:- a reference input (11) for receiving a first reference signal (R1) representative of far-end speech (RX) from the communication network (4);- a microphone input (12) for receiving a first microphone signal (M1) from the microphone (8) comprising a near-end speech portion representative of near-end speech (SN) from the local user (1) and an echo portion representative of direct and/or indirect acoustic feedback (SE) of far-end speech (SF) from the loudspeaker (7) to the microphone (8);- a first double-talk detector (13) configured to estimate a first double-talk likelihood (L1) being a likelihood of simultaneous presence of near-end speech and far-end speech in the first microphone signal (M1) and to provide a first double-talk signal (D1) indicating the first double-talk likelihood (L1);- a first echo reduction block (14) connected to receive a second reference signal (R2) equal to or derived from the first reference signal (R1), a second microphone signal (M2) equal to or derived from the first microphone signal (M1), and the first double-talk signal (D1), wherein the first echo reduction block (14) is configured to adaptively process the second microphone signal (M2) based on the second reference signal (R2) to estimate a first echo-cancelled signal (C1), and wherein the first echo reduction block (14) is further configured to control the processing of the second microphone signal (M2) in dependence on the first double-talk likelihood (L1); and- a near-end speech output (15) connected to receive a first echo-suppressed signal (S1) equal to or derived from the first echo-cancelled signal (C1), wherein the near-end speech output (15) is configured to provide a near-end speech signal (TX) comprising near-end speech for transmission to the communication network (4) based on the first echo-suppressed signal (S1).The echo suppressor (10) is characterized in that:- the echo suppressor (10) is configured to estimate a third microphone signal (M3) based on the first microphone signal (M1) and the first reference signal (R1); and- the first double-talk detector (13) is connected to receive the third microphone signal (M3) and is further configured to estimate the first double-talk likelihood (L1) in dependence on the first reference signal (R1) and the third microphone signal (M3).By estimating the double-talk likelihood (L1) in dependence on the third microphone signal (M3) wherein, preferably, the echo portion is reduced relative to the near-end speech portion when compared with the first microphone signal (M1), the first double-talk detector (13) may estimate the first double-talk likelihood (L1) with better accuracy and/or reliability and thereby enable the first echo reduction block (14) and the echo suppressor (10) to further reduce the echo portion relative to the near-end speech portion in the near-end speech signal (TX).

Anmelder

Firma

GN Hearing A/S

Land

🇩🇰 Dänemark

Fachgebiete