18540361. Method and Device for Voice Activity Detection simplified abstract (Telefonaktiebolaget LM Ericsson (publ))

From WikiPatents
Jump to navigation Jump to search

Method and Device for Voice Activity Detection

Organization Name

Telefonaktiebolaget LM Ericsson (publ)

Inventor(s)

Martin Sehlstedt of Luleå (SE)

Method and Device for Voice Activity Detection - A simplified explanation of the abstract

This abstract first appeared for US patent application 18540361 titled 'Method and Device for Voice Activity Detection

Simplified Explanation

The abstract describes a method and apparatus for voice activity detection (VAD) that involves creating a signal indicative of a primary VAD decision, determining hangover addition based on short term and/or long term activity measures, and creating a signal indicative of a final VAD decision.

  • Voice Activity Detection (VAD) method and apparatus
  • Signal creation for primary VAD decision
  • Determination of hangover addition based on activity measures
  • Signal creation for final VAD decision

Potential Applications

The technology can be applied in:

  • Speech recognition systems
  • Telecommunications for noise reduction
  • Voice-controlled devices

Problems Solved

  • Accurate detection of voice activity in noisy environments
  • Efficient utilization of resources by activating voice processing only when necessary

Benefits

  • Improved accuracy in voice recognition
  • Enhanced user experience with voice-controlled devices
  • Reduction in background noise interference

Potential Commercial Applications

  • Mobile phones and smart speakers
  • Call centers for improved call quality
  • Automotive systems for hands-free operation

Possible Prior Art

One possible prior art in voice activity detection is the use of energy-based VAD algorithms that analyze the energy levels in audio signals to detect voice activity.

Unanswered Questions

How does the method handle variations in speech patterns and accents?

The article does not address how the VAD method adapts to different speech patterns and accents, which could affect its accuracy in detecting voice activity.

What is the computational complexity of the VAD algorithm?

The article does not provide information on the computational resources required to implement the VAD algorithm, which could impact its feasibility for real-time applications.


Original Abstract Submitted

In accordance with an example embodiment of the present invention, disclosed is a method and an apparatus for voice activity detection (VAD). The VAD comprises creating a signal indicative of a primary VAD decision and determining hangover addition. The determination on hangover addition is made in dependence of a short term activity measure and/or a long term activity measure. A signal indicative of a final VAD decision is then created.