Skip to content
TechSpec
  • Foreword
  • 1 Scope
  • 2 References
  • 3 Technical Description of VAD Option 1
    • 3.1 Definitions, symbols and abbreviations
    • 3.2 General
    • 3.3 Functional description
      • 3.3 Functional description
      • 3.3.1 Filter bank and computation of sub-band levels
      • 3.3.2 Pitch detection
      • 3.3.3 Tone detection
      • 3.3.4 Correlated Complex Signal Analysis (and detection)
      • 3.3.5 VAD decision
  • 4 Technical Description of VAD Option 2
    • 4.1 Definitions, symbols and abbreviations
    • 4.2 General
    • 4.3 Functional description
      • 4.3 Functional description
      • 4.3.1 Frequency Domain Conversion
      • 4.3.2 Channel Energy Estimator
      • 4.3.3 Channel SNR Estimator
      • 4.3.4 Voice Metric Calculation
      • 4.3.5 Frame SNR and Long-Term Peak SNR Calculation
      • 4.3.6 Negative SNR Sensitivity Bias
      • 4.3.7 VAD Decision
      • 4.3.8 Spectral Deviation Estimator
      • 4.3.9 Sinewave Detection
      • 4.3.10 Background Noise Update Decision
      • 4.3.10 Background Noise Estimate Update
  • 5 Computational details

4.2 General

26.0943GPPAdaptive Multi-Rate (AMR) speech codecMandatory speech CODEC speech processing functionsRelease 17TSVoice Activity Detector (VAD)

Tools: ARFCN - Frequency Conversion for 5G NR/LTE/UMTS/GSM

The function of the VAD algorithm is to indicate whether each 20 ms frame contains signals that should be transmitted, i.e. speech, music or information tones. The output of the VAD algorithm is a Boolean flag (VAD_flag) indicating presence of such signals.