Conference Call Speaker Identification via Directional Microphones
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In conference calls involving multiple participants across different locations, remote participants often struggle to identify who is speaking among local participants in a physical conference room, as there is no clear visual or audio indication beyond voice recognition, which can be challenging with multiple speakers and new attendees joining.
Innovation Solution
An electronic device system that uses user profiles, transmitting devices, and directional microphones to associate audio signals with participant identifiers, such as avatars, and transmit this information to remote participants, ensuring accurate identification of speakers through proximity detection and authentication.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple local participants share a common speaker phone for communicating with virtual participants, then the conference call system can support multiple participants, but remote participants cannot identify who is speaking among local participants
Solution Approach 1:
The patent segments the audio signal processing by assigning separate directional microphones to different local participants. Each participant has their own audio capture device that spatially isolates their voice, allowing the system to identify and transmit speaker information to remote participants without requiring them to manually indicate their identity.
Solution Approach 2:
The patent introduces an intermediary system that automatically links audio signals from directional microphones to participant identifiers. This intermediary processing layer matches spoken words to specific participants based on audio source identification, eliminating the need for remote participants to manually identify speakers while maintaining support for multiple local participants.
2Device complexity
If remote participants rely on voice recognition to identify local participants, then no additional identification infrastructure is needed, but identification becomes challenging with multiple speakers and new attendees
Solution Approach 1:
The patent replaces manual voice recognition with an automated audio source identification system using directional microphones. Instead of relying on remote participants to manually identify speakers through voice recognition, the system automatically detects the direction and source of audio signals, linking them to specific participants based on their spatial position and pre-configured participant information.
3Loss of information
If local participants must state their names prior to speaking to enable identification, then speaker identification is possible, but communication efficiency decreases due to repeated introductions
Solution Approach 1:
The patent performs preliminary action by pre-configuring participant information and assigning directional microphones to specific participants before the conference call begins. This advance setup allows the system to automatically identify speakers based on audio source location without requiring participants to repeatedly state their names during the call, thereby maintaining speaker identification accuracy while improving communication efficiency.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables remote participants to clearly identify who is speaking in real-time by displaying participant avatars and audio signals simultaneously, improving communication clarity and security by authenticating participants through unique identifiers.
Implementation Method 1
at least one directional microphone receives the audio signal
Data Source
Figure 1
Figure 2~4
Figure 5
AI summary
An example electronic device includes a memory to store a user profile including first identifying information of a first participant in a conference call. A receiver receives second identifying information of the first participant from a transmitting device associated with the first participant, wherein the first identifying information and the second identifying information form an identifier for the first participant. An audio encoder receives an audio signal. A processor, in response to determining which transmitting device is nearest to a source of the audio signal relative to other transmitting devices, identifies the first participant as a source of the audio signal, and combines the identifier for the first participant with the audio signal generated by the first participant. A router forwards the combined identifier and audio of the first participant to a receiving device of a second participant in the conference call.