Wireless Signal Sharing for Speech Recognition Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In group conversations, speech recognition technology faces challenges with poor audio signal quality due to distance from the microphone, leading to suboptimal transcription for individuals with hearing difficulties.
Innovation Solution
A system and method for sharing microphone signals among wireless devices, allowing users to form a signal sharing group, where devices can exchange and process audio signals for improved speech recognition and closed captioning, with options for selecting the best signal source and performing signal processing adaptations for hearing aids.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single microphone is used for speech recognition in group conversations, then the device complexity is low, but the speech recognition accuracy deteriorates when speakers are far from the microphone
Solution Approach 1:
The patent combines multiple microphones from different portable devices into a unified signal sharing group, allowing the system to capture speech from multiple spatial locations simultaneously. This merging of microphone resources enables accurate speech recognition even when speakers are distant from any single microphone, while distributing the processing load across multiple devices to manage complexity.
Solution Approach 2:
The patent creates a universal signal sharing framework where portable devices can function both as microphones and as speech recognition processors. Each device in the signal sharing group can capture audio and process speech recognition, allowing any device to serve multiple functions depending on the conversation dynamics, thereby improving accuracy without proportionally increasing overall system complexity.
2Measurement precision
If manual microphone passing is implemented to improve signal quality, then the speech recognition accuracy improves, but the ease of operation deteriorates and time is lost
Solution Approach 1:
The patent replaces the mechanical approach of physically passing microphones between speakers with a wireless signal sharing system. Multiple microphones simultaneously capture speech from different positions, and the system electronically selects or combines the best signals, eliminating the need for manual microphone handling while maintaining high audio quality and improving convenience.
Solution Approach 2:
The system pre-establishes signal sharing connections between multiple portable devices before the conversation begins. This preliminary setup ensures that microphones are already positioned optimally for each speaker, and the speech recognition system is ready to process signals from any device, eliminating the need for real-time adjustments or manual repositioning during the conversation.
3Measurement precision
If multiple devices process speech recognition simultaneously, then the speech recognition accuracy improves, but the use of energy increases
Solution Approach 1:
The patent implements selective speech recognition processing where not all devices in the signal sharing group process every speech segment. Instead, devices process speech based on their proximity to the speaker, signal quality metrics, or workload distribution. This partial processing approach maintains high recognition accuracy by utilizing multiple perspectives when needed, while reducing overall energy consumption by avoiding redundant processing across all devices for every utterance.
Data Source
AI summary
Methods and devices are described for allowing users to use portable computer devices such as smart phones to share microphone signals and/or closed captioning text generated by speech recognition processing of the microphone signals. Under user direction, the portable devices exchange messages to form a signal sharing group to facilitate their conversation.