Granted Patent B2

US 12,542,149 · App. 17/666,645 · Granted Feb 3, 2026

Inventor: Benjamin Bernard (Karlsruhe, DE)

Assignee: DR. ING. H.C. F. PORSCHE AKTIENGESELLSCHAFT

G10L21/0364G06V40/176G10L15/25G10L21/0232G10L25/84

Abstract

Claims (25)

1 . A method ( 100 ) for improving speech intelligibility in a passenger compartment of a motor vehicle, comprising:

using a seat occupancy detection apparatus to identify where in a passenger compartment passengers are seated;

detecting ( 110 ) an active speaker, or speech, in the passenger compartment;

identifying ( 120 ) a voice frequency profile for the active speaker;

identifying where in the passenger compartment the active speaker is located based on the seat occupancy detection apparatus and further based on at least one of a camera or a microphone;

influencing ( 130 ) at least one apparatus for producing sound in the passenger compartment in such a way that a level of sound emitted into the passenger compartment by the apparatus is reduced at frequencies or in frequency ranges that coincide with or are adjacent to the frequencies, or frequency ranges, of the speaker that are relevant to speech intelligibility; and

altering acoustic features or parameters of at least one structure, panelling, linings, cladding or coverings in the passenger compartment in such a way that sound transmitted into the passenger compartment of the motor vehicle is reduced or damped by this altering of the structure or panelling at the frequencies or in the frequency ranges that are relevant to the speech intelligibility of the speaker, or at frequencies or in frequency ranges that are adjacent to said frequencies or frequency ranges, by means of this altering of the at least one structure, panelling, linings, cladding or coverings in the passenger compartment.

2 . The method of claim 1 , wherein the at least one of a camera or a microphone comprises at least one microphone, and the detecting ( 110 ) of an active speaker comprises:

using the at least one microphone in the passenger compartment for detecting frequencies and/or sounds that are typical of human speech and for confirming where in the passenger compartment the frequencies or sounds that are typical of human speech are coming from.

3 . The method of claim 1 , wherein the at least one of a camera or a microphone comprises at least one camera, and the detecting ( 110 ) of an active speaker comprises:

using the at least one camera in the passenger compartment for detecting facial movements of persons that are typical of speech and for confirming where in the passenger compartment the active speaker is located.

4 . The method of claim 1 , wherein the at least one of a camera or a microphone comprises at least one microphone, and identifying ( 120 ) a voice frequency profile for the active speaker comprises:

comparing speech recorded by at least one microphone arranged in the passenger compartment with speech profiles of persons that are obtained from a memory.

5 . The method of claim 1 , wherein the at least one of a camera or a microphone comprises at least one camera, and identifying ( 120 ) the active speaker comprises:

comparing a face recorded by at least one camera arranged in the passenger compartment with representations of faces for which voice profiles are stored in a memory.

6 . The method of claim 1 , wherein the detection of an active speaker, or detection of speech, and/or the identification of the speaking individual further involves evaluation of additional signals provided by sensors arranged in the passenger compartment.

7 . The method of claim 1 , wherein the at least one apparatus for producing sound in the passenger compartment is influenced in such a way that a level of the sound emitted by the apparatus is reduced, and/or sound is reduced by destructive interference at least at locations in the passenger compartment at which the seat occupancy detection apparatus has identified a passenger other than the active speaker.

8 . The method of claim 1 , wherein the at least one of a camera or a microphone comprises at least one microphone, and the method further comprising delayed and/or correct-phase reproduction of speech recorded by the at least one microphone in the passenger compartment by way of an apparatus for producing sound that is arranged in the passenger compartment.

9 . The method of claim 8 , wherein the delayed and/or correct-phase reproduction of speech is performed in one or more areas of the passenger compartment in which the seat occupancy detection apparatus has identified a passenger.

10 . The method of claim 1 , further comprising:

determining a voice frequency profile of a speaker,

ascertaining frequencies, or frequency ranges, that are important to the speech intelligibility of the speaker,

determining features of the face of the speaker that are suitable for identifying the speaker and typical facial movements of the speaker when speaking, and

storing in a memory the ascertained voice frequency profile and/or frequencies, or frequency ranges, important to the speech intelligibility of the speaker and the features of the face of the speaker that are suitable for identifying the speaker and the typical facial movements of the speaker when speaking, wherein the memory can be accessed by an apparatus for improving speech intelligibility in the passenger compartment, and wherein the apparatus is arranged in or associated with the passenger compartment.

11 . The method of claim 1 , wherein the at least one of a camera or a microphone comprises at least one microphone and wherein the step of influencing ( 130 ) at least one apparatus for producing sound in the passenger compartment comprises reproducing the speech detected by the microphone while simultaneously reducing, by destructive interference sound that is not identified as speech.

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (24)