Speech Reproduction Control for Distinguishing Limited Conversations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In existing online conference systems, participants who instruct for limited conversation often struggle to distinguish between regular and limited conversation speech, leading to difficulty in identifying and managing such conversations effectively.
Innovation Solution
A speech reproduction control system that includes a speech reproducer and a character reproducer, allowing for parallel reproduction of speech data and conversion into characters, with the option to mute or reduce the volume of certain speech data to manage conversations effectively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If speech data is reproduced only in speech form, then the reproduction method is simple, but it becomes difficult to distinguish between limited conversation and regular conversation
Solution Approach 1:
The speech data reproduction is segmented into two distinct channels: speech reproduction and character reproduction. Limited conversation data is reproduced as characters while regular conversation data is reproduced as speech, allowing participants to clearly distinguish between the two types of conversations through the different reproduction forms.
Solution Approach 2:
The system adds a new dimension to speech reproduction by introducing character reproduction as a separate output modality. This dimensional change allows the system to convey not only the content of the speech but also the type of conversation (limited or regular) through the choice of reproduction form.
2Productivity
If multiple speech data are reproduced in parallel in speech form, then all speech data can be heard, but it becomes difficult to identify which speech is limited conversation
Solution Approach 1:
Different local qualities are applied to different speech data streams: limited conversation speech data is converted to characters for display, while regular conversation speech data is reproduced as speech. This local differentiation in reproduction quality enables participants to easily identify which speech corresponds to limited conversation among multiple parallel speech streams.
3Object-affected harmful factors
If speech volume is reduced for all speech data, then background speech can be heard, but limited conversation cannot be distinguished from regular conversation
Solution Approach 1:
The system segments the speech data into limited conversation and regular conversation, applying different reproduction strategies to each segment. This segmentation allows the system to manage background speech interference while preserving the ability to distinguish conversation types through the use of character reproduction for limited conversations.
Data Source
AI summary
A speech reproduction control system includes a speech reproducer that reproduces speech data in speech, a character reproducer that converts the speech data into characters for reproduction, and a hardware processor that reproduces first speech data and second speech data in parallel, wherein the hardware processor reproduces the first speech data by using the character reproducer, turns off reproduction of the first speech data by the speech reproducer or reproduces the first speech data at a volume lower than a volume for reproduction of the second speech data, and reproduces the second speech data by using the speech reproducer.