US 2026/0301742 A1Published application

Speech transcription by an electronic device on visual detection of a participant side conversation during a communicaton session

AssigneeMotorola Mobility LLC

InventorsRANJEET GUPTA, RAHUL BHARAT DESAI

Life of this patent

Estimating the term…

Loading patent…

Abstract

An electronic device, a method and a computer program product for transcribing speech from a communication session to text in response to detecting a local participant in a side conversation. The method includes identifying an audio transmission feature to the communication session of the local participant is set to a muted mode and capturing a preview image stream. The method includes determining if the preview image stream includes facial images that correspond to faces of the local participant a non-participant. The method includes determining, based on facial movements of the local participant, if the local participant is engaged in a side conversation with the non-participant. The method includes detecting speech input received via the communication session from at least one remote participant. The method includes generating a text transcript or summary of the speech input and visually presenting the text transcript or summary to the local participant.

Claims20

Dependent claims

An electronic device comprising:

a communications subsystem that enables the electronic device to communicatively connect with at least one second electronic device via a communication session;
at least one camera including a first camera;
a first audio input device;
a video controller that presents video output;
a memory having stored thereon a communication module and a communication session speech to text (CSST) module for transcribing speech received from the communication session into text;
and
at least one processor communicatively coupled to each of the communications subsystem, the at least one camera, the first audio input device, the video controller, and the memory, and which executes program code of the communication module and the CSST module, the at least one processor configured to cause the electronic device to:
during a first communication session between a local participant and one or more remote participants, identify if an audio transmission feature of the local participant to the first communication session is set to a muted mode, wherein during the muted mode, audio input detected by the first audio input device is not presented to the first communication session;
capture a first preview image stream via the first camera;
identify if at least two facial images are included within the first preview image stream;
in response to identifying that the at least two facial images are included within the first preview image stream, determine if the at least two facial images include a first facial image that corresponds to a first face of the local participant and a second facial image that corresponds to a second face of a non-participant in a vicinity of the local participant;
in response to determining that the at least two facial images include the first facial image that corresponds to the first face of the local participant and the second facial image that corresponds to the second face of the non-participant, determine, based on facial movements of the local participant, if the local participant is engaged in a side conversation with the non-participant;
and
in response to determining that the local participant is engaged in the side conversation with the non-participant:
detect speech input received via the first communication session from at least one remote participant;
generate at least one of a text transcript and a summary of the speech input;
and
visually present the text transcript or summary to the local participant.

15 more elements

A method comprising:

during a first communication session between a local participant and one or more remote participants, identifying, via at least one processor of an electronic device, if an audio transmission feature of the local participant to the first communication session is set to a muted mode, wherein during the muted mode, audio input detected by a first audio input device is not presented to the first communication session;
capturing a first preview image stream via a first camera;
identifying if at least two facial images are included within the first preview image stream;
in response to identifying that the at least two facial images are included within the first preview image stream, determining if the at least two facial images include a first facial image that corresponds to a first face of the local participant and a second facial image that corresponds to a second face of a non-participant in a vicinity of the local participant;
in response to determining that the at least two facial images include the first facial image that corresponds to the first face of the local participant and the second facial image that corresponds to the second face of the non-participant, determining, based on facial movements of the local participant, if the local participant is engaged in a side conversation with the non-participant;
and
in response to determining that the local participant is engaged in the side conversation with the non-participant:
detecting speech input received via the first communication session from at least one remote participant;
generating at least one of a text transcript and a summary of the speech input;
and
visually presenting the text transcript or summary to the local participant.

8 more elements

A computer program product comprising:

a computer readable storage device having stored thereon program code which, when executed by at least one processor of an electronic device having a communications subsystem that enables the electronic device to communicatively connect with at least one second electronic device via a communication session, at least one camera including a first camera, a first audio input device, and a video controller that presents video output, configures the electronic device to complete the functionality of:
during a first communication session between a local participant and one or more remote participants, identifying if an audio transmission feature of the local participant to the first communication session is set to a muted mode, wherein during the muted mode, audio input detected by the first audio input device is not presented to the first communication session;
capturing a first preview image stream via the first camera;
identifying if at least two facial images are included within the first preview image stream;
in response to identifying that the at least two facial images are included within the first preview image stream, determining if the at least two facial images include a first facial image that corresponds to a first face of the local participant and a second facial image that corresponds to a second face of a non-participant in a vicinity of the local participant;
in response to determining that the at least two facial images include the first facial image that corresponds to the first face of the local participant and the second facial image that corresponds to the second face of the non-participant, determining, based on facial movements of the local participant, if the local participant is engaged in a side conversation with the non-participant;
and
in response to determining that the local participant is engaged in the side conversation with the non-participant:
detecting speech input received via the first communication session from at least one remote participant;
generating at least one of a text transcript and a summary of the speech input;
and
visually presenting the text transcript or summary to the local participant.

9 more elements

Dependencies are inferred from the claim text.

Facts

Application: 19/094,958Filed 30 Mar 2025
Published: 1 Oct 2026
Term ends: Not estimatedThe source records don’t support an estimate.

Classification7 codes

G10L 15/26
Speech recognition: speech to text systems
G06V 40/161
Human faces, e.g. facial parts, sketches or expressions: detection; Localisation; Normalisation
G06V 40/172
Human faces, e.g. facial parts, sketches or expressions: classification, e.g. identification
G06V 40/20
Recognition of biometric, human-related or animal-related patterns in image or video data: movements or behaviour, e.g. gesture recognition

Family

No related US applications are recorded in the indexed filing history.

Timeline

Citations

Open elsewhere

Building an agent or pulling many patents? Use the API or MCP server instead of scraping pages. Get an API key