Earphone Voice Recognition Audio Buffering

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Voice recognition systems, such as those in earphones, often experience word loss and unreliable recognition due to the time delay between detecting mouth movement and picking up sound, leading to incomplete wake-word detection.

Innovation Solution

A method for voice recognition in earphones that buffers and combines audio data received before and after a mouth movement is detected, ensuring complete wake-word capture by defining specific durations for data reception and processing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the voice recognition system waits to detect mouth movement before recording audio data, then the recognition accuracy is improved, but words will be lost during the detection time delay

Engineering Contradiction:
Improvevoice recognition accuracyVSAvoidword loss
Core Design Contradiction:
Measurement precisionVSLoss of information

Solution Approach 1:

The system performs preliminary action by starting to buffer audio data before the mouth movement detection is complete. When the detection triggers, the audio data has already been captured during the detection period, preventing any word loss while maintaining recognition accuracy.

Inventive Principle:
Principle #10Preliminary action

2Loss of information

If the system starts recording immediately without waiting for mouth movement detection, then word loss is prevented, but recognition reliability decreases due to incomplete wake-word detection

Engineering Contradiction:
Improveword loss preventionVSAvoidrecognition reliability
Core Design Contradiction:
Loss of informationVSReliability

Solution Approach 1:

The system performs preliminary action by starting to buffer audio data before the mouth movement detection is complete. When the detection triggers, the audio data has already been captured during the detection period, preventing any word loss while maintaining recognition accuracy.

Inventive Principle:
Principle #10Preliminary action

3Loss of information

If the system buffers audio data during the detection period, then word loss is prevented, but the device complexity increases

Engineering Contradiction:
Improveword loss preventionVSAvoidsystem complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The system merges the audio buffering function with the existing detection mechanism. The audio data buffered during the detection period is seamlessly integrated with the recognition process, preventing word loss without requiring a separate complex subsystem.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentEP3764352B1Method for voice recognition via earphone and earphone
Publication Date: 2023.05.31 GUANGDONG OPPO MOBILE TELECOMMUNICATIONS CORP LTD
  • EP3764352B1 patent drawingFigure 1~2
  • EP3764352B1 patent drawingFigure 3~4
  • EP3764352B1 patent drawingFigure 5~6

AI summary

A method for voice recognition via an earphone is disclosed. The method includes receiving first audio data via the first microphone and buffering the first audio data in response to the first trigger signal; receiving second audio data via the first microphone and recognizing whether the first audio data contains data of a wake-on-voice word in response to the second trigger signal; and recognizing whether the second audio data contains data of the wake-on-voice word. The first audio data is received and buffered in a first duration starting from when the first trigger signal is received and ending when the second trigger signal is received. The second audio data is received in a second duration starting from when the second trigger signal is received and ending when whether the first audio data contains data of the wake-on-voice word is recognized.