Voice Recognition Processor Phoneme Detection for Continuous Speech

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing electronic devices struggle to detect a call command for speech recognition services when the user utters consecutive voices containing the call command without silences, leading to an unnatural user interface.

Innovation Solution

An electronic device equipped with a processor that analyzes user voice sections in phonemic units to detect wake-up words or predetermined words, including pronouns, allowing for natural utterance and continuous speech recognition service activation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the speech recognition service is called only when the call command is between silences, then the service can be accurately triggered, but the user interface becomes unnatural and inconvenient for continuous speech

Engineering Contradiction:
Improvecall command recognition accuracyVSAvoiduser interface naturalness
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The system dynamically adjusts its detection mode based on the continuity of speech. It transitions from silence-based detection for isolated commands to phoneme-based pattern recognition for continuous speech, allowing the detection mechanism to adapt to different user behaviors and speech patterns

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes the detection parameters from requiring silence boundaries to accepting phoneme sequences within a time window. This parameter change allows the system to recognize call commands embedded in continuous speech by analyzing phoneme patterns rather than relying on silence delimiters

Inventive Principle:
Principle #35Parameter changes

2Reliability

If the user must utter the call command for each speech recognition service call, then the service can be reliably activated, but the user experience deteriorates when pronouns or natural language are used

Engineering Contradiction:
Improveservice activation reliabilityVSAvoidlanguage usage flexibility
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The phoneme-based detection system serves multiple functions: it detects explicit call commands, recognizes pronouns referring to the assistant, and identifies natural language triggers. This universal detection mechanism handles various language patterns uniformly, making the system adaptable to different user expressions

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The phoneme sequence analysis acts as an intermediary between raw audio input and service activation. Instead of directly requiring complete call commands, the system uses phoneme patterns as a mediator to indirectly detect user intent, allowing pronouns and natural language to trigger services

Inventive Principle:
Principle #24Intermediary (Mediator)

3Measurement precision

If consecutive voice utterances are analyzed in detail to detect call commands, then detection accuracy improves, but the processing time and computational complexity increase

Engineering Contradiction:
Improvecall command detection accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs partial analysis by focusing only on phoneme sequences within a predetermined time window rather than analyzing entire continuous speech. This partial action approach detects call commands without processing all audio data, reducing time loss while maintaining detection accuracy

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3796316B1Electronic device and control method thereof
Publication Date: 2025.03.05 SAMSUNG ELECTRONICS CO LTD
  • EP3796316B1 patent drawingFigure 1
  • EP3796316B1 patent drawingFigure 2
  • EP3796316B1 patent drawingFigure 3A

AI summary

An electronic device of the present disclosure comprises: a communication unit; a memory; and a processor for: detecting a voice section in an audio signal acquired by the electronic device; identifying whether a wake-up word stored in the memory exists in a user voice included in the detected voice section; when it is identified that the wake-up word exists in the user voice, transmitting, via the communication unit, the user voice to a server for providing a voice recognition service; and when response information for the user voice is received from the server, providing a response to the user voice on the basis of the received response information, wherein the processor identifies that the wake-up word exists in the user voice, when a part of the user voice matches the wake-up word. In particular, a method for acquiring a natural language for providing a response may use an artificial intelligence model learned according to at least one of machine learning, a neural network, and a deep learning algorithm.