Bluetooth Headset Speech Recognition via Lossless Audio Link

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Bluetooth headsets face challenges in speech recognition due to the lossy nature of existing Bluetooth profiles (HFP and HSP) for voice transmission, which degrades speech recognition accuracy and introduces packet loss, and require complex electronics or constant cloud connectivity.

Innovation Solution

Establishing a dual wireless data link system in wearable devices, using a lossless Bluetooth link for speech recognition and a lossy link for voice calls, with custom profiles and compression techniques to improve data integrity and reduce noise interference.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If cloud-based recognition systems are used, then speech recognition capability is provided, but requirement of an always-on connection to the cloud increases device complexity and energy consumption

Engineering Contradiction:
Improvespeech recognition capabilityVSAvoidconnection requirement
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent segments the speech recognition system into two parts: local speech recognition processing embedded in the Bluetooth headset and cloud-based services. This allows the headset to perform basic speech recognition independently without requiring constant cloud connection, while still enabling advanced features when connected.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements self-service by embedding speech recognition functionality directly in the Bluetooth headset, enabling it to process speech commands locally without external assistance. This reduces dependency on cloud services and allows the device to function autonomously for basic speech recognition tasks.

Inventive Principle:
Principle #25Self-service

2Ease of operation

If HFP or HSP profiles are used for audio transmission, then voice communication is enabled, but packet loss degrades speech recognition accuracy

Engineering Contradiction:
Improvevoice communicationVSAvoidspeech recognition accuracy
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The patent changes the transmission parameters by using a 16 kHz sampling rate for speech recognition audio transmission, which is double the standard 8 kHz rate. This higher sampling rate provides more detailed speech information that can tolerate packet loss better, maintaining speech recognition accuracy even when using lossy transmission profiles like HFP or HSP.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent transmits more audio data than strictly necessary by using a higher sampling rate (16 kHz vs 8 kHz). This excessive data transmission ensures that even if some packets are lost during transmission, sufficient speech information remains to maintain accurate speech recognition.

Inventive Principle:
Principle #16Partial or excessive action

3Productivity

If speech recognition service runs on the Bluetooth headset itself, then local processing is achieved, but complex and expensive electronics are required in the headset

Engineering Contradiction:
Improvelocal speech processingVSAvoidelectronics complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent uses a cost-effective approach by implementing a simplified speech recognition engine in the Bluetooth headset that works with higher sampling rate audio data. This lighter-weight processing approach avoids the need for expensive, complex electronics while still providing functional speech recognition capability.

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

Data Source

PatentEP3090531B1Smart bluetooth headset for speech command
Publication Date: 2019.04.10 KOPIN CORP
  • EP3090531B1 patent drawingFigure 1
  • EP3090531B1 patent drawingFigure 2
  • EP3090531B1 patent drawingFigure 3

AI summary

A method of interfacing with a serving device from a wearable device worn by a user, the method includes establishing a lossless and wireless data link between the serving device and the wearable device. The method further includes collecting, by the wearable device, audio data from one or more microphones of the wearable device. The method also includes sending, by the wearable device, the collected audio data to the serving device through the lossless and wireless data link. The method may further include providing, by the serving device, speech recognition services associated with the audio data.