Speech Processing Device for Silent Call Partner Identification

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Users cannot easily identify call partners by hearing alone, especially when call partners are silent, as existing techniques fail to provide distinct auditory cues for silent participants, leading to a lack of motivation for users to perform operations to output utterer information.

Innovation Solution

A speech processing device that includes a call partner identification unit, a background sound selection unit, and a synthesis unit, which identifies call partners and selects relevant background sounds to synthesize with call speech signals, allowing users to differentiate call partners through distinct auditory cues.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If background sounds are synthesized with call speech signals to enable call partner identification by hearing alone, then call partner identifiability is improved, but device complexity increases

Engineering Contradiction:

Improvecall partner identifiabilityVSAvoidprocessing system complexity

Core Design Contradiction:

Loss of informationVSDevice complexity

Solution Approach 1:

Background sounds are pre-associated with each call partner and stored in the system before the actual call occurs. When a call partner is identified, the corresponding background sound is automatically selected and synthesized with the speech signal, eliminating the need for complex real-time analysis and reducing processing complexity during the call.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

Background sounds serve as an intermediary element that bridges the gap between visual display information and auditory-only identification. By introducing this intermediate audio layer, the system enables call partner identification through hearing alone without requiring direct visual contact or complex speech analysis algorithms.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Ease of operation

If distinct auditory cues are provided for each call partner through background sounds, then user experience is improved, but information processing requirements increase

Engineering Contradiction:

Improveuser experienceVSAvoidaudio data processing

Core Design Contradiction:

Ease of operationVSQuantity of substance

Solution Approach 1:

The audio output is segmented into distinct components: the primary speech signal from the call partner and the secondary background sound associated with that call partner. This segmentation allows each component to be processed and managed independently, reducing the overall processing burden while providing rich auditory cues for user identification.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Different background sounds are assigned to different call partners, creating local quality variations in the audio output. Each call partner has their own distinctive background sound characteristics, allowing users to easily differentiate between multiple participants without requiring complex processing of the entire audio stream.

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS20240386874A1Speech processing device, speech processing method, and recording medium

Publication Date: 2024.11.21 NEC CORP

AI summary

A call partner identification means identifies a call partner in order to make it possible for a user to easily identify the call partner by only the sense of hearing. A background sound selection means selects a background sound corresponding to the identified call partner. A synthesis means synthesizes a call speech signal and the selected background sound.