Automatic Volume Control for Captioning Communication Services
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Hearing-impaired individuals face challenges in communication due to echo issues during captioning communication sessions, where hybrid echo and acoustic echo can hinder the understanding of conversations, and existing echo cancellation systems are not always effective, especially in environments with impedance imbalances.
Innovation Solution
A communication device and method that automatically adjusts the volume of the audio stream based on the active talker situation by comparing near-end and far-end voice signals, using an echo modifier to distort the echo portion of the far-end voice signal, allowing the captioning communication service to better distinguish between the far-end voice and echo, thereby improving transcription accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If conventional echo cancellation systems are used, then hybrid echo and acoustic echo can be reduced, but transcription accuracy deteriorates when impedance imbalances are present
Solution Approach 1:
Instead of attempting to cancel the echo through conventional means, the patent inverts the approach by deliberately adding distortion to the echo signal. This distorted echo is then subtracted from the received signal, effectively isolating and enhancing the far-end voice component for more accurate transcription by the captioning service.
Solution Approach 2:
The patent changes the parameters of the echo signal by applying distortion transformations. By modifying the echo signal's characteristics through distortion and then subtracting it, the system transforms the echo from a harmful interference into a useful component that helps isolate the desired voice signal.
2Productivity
If the call assistant listens to the far-end user's audio signal to generate captions, then text captions can be provided to hearing-impaired users, but echo interference reduces the accuracy of voice recognition
Solution Approach 1:
The patent converts the harmful echo into a beneficial element by using it as a reference signal. The distorted echo, when subtracted from the received signal, helps isolate the far-end voice, thereby transforming what was originally interference into a tool that improves voice recognition accuracy for the captioning service.
Solution Approach 2:
The patent introduces distortion as an intermediary transformation process. By applying distortion to the echo signal before subtraction, it creates an intermediate distorted echo signal that serves as a mediator to more effectively separate the echo component from the desired voice signal, improving the captioning service's ability to accurately transcribe speech.
3Ease of operation
If automatic volume control is implemented based on active talker detection, then user experience is improved, but system complexity increases
Solution Approach 1:
The patent implements a self-service volume control system that automatically detects which user is actively speaking and adjusts the volume accordingly without requiring manual user intervention. The system monitors the audio signals, identifies the active talker, and autonomously controls the volume, providing an improved user experience while keeping the control mechanism relatively simple.
Solution Approach 2:
The patent employs feedback mechanisms where the system continuously monitors the audio signals from both near-end and far-end users, detects the active talker based on signal presence, and adjusts the volume control in response. This feedback loop enables automatic adaptation to the communication situation, improving ease of operation through intuitive volume management.
Data Source
AI summary
Apparatuses and methods are disclosed for automatic volume control of an audio stream reproduced by a captioning communication service for use by a call assistant in generating a text transcription of a communication session between a hearing-impaired user and a far-end user. The automatic volume control automatically adjusts a volume of the audio stream reproduced by the captioning communication service responsive to a volume control command identifying which of the far-end voice signal and the near-end voice signal is active at a given time. The system further includes an echo modifier configured to add distortion to an echo portion of the far-end voice signal when generating the audio stream.