SBAA600 October   2024 TAA5212 , TAC5111 , TAC5112 , TAC5211 , TAC5212

 

  1.   1
  2.   Abstract
  3.   Trademarks
  4. 1Introduction
  5. 2Voice Activity Detection
    1. 2.1 VAD Configurations
      1. 2.1.1 User, Auto, Intermediate
      2. 2.1.2 VAD With ADC Recording
      3. 2.1.3 VAD Monitoring Channel
      4. 2.1.4 VAD Interrupt Pin
      5. 2.1.5 MICBIAS Enable During PDM Monitoring
      6. 2.1.6 VAD Clock Configurabilty
    2. 2.2 VAD Parameters
      1. 2.2.1 Initial Learning Period
      2. 2.2.2 Hold Over Counter
      3. 2.2.3 Wakeup Wait
      4. 2.2.4 Threshold
  6. 3VAD Performance Results
  7. 4Examples
  8. 5Summary
  9. 6References

2 Voice Activity Detection

The VAD algorithm uses a decision tree classification-based algorithm for voice activity detection. The VAD block monitors the input signal from the microphone channels for a voice-like profile and upon detection of a voice activity pattern, the VAD block triggers an interrupt. The VAD monitors for both an onset of voice activity as well as the end of voice activity. Both of these events can be mapped to interrupts.

The TAx511x or TAx521x device also has the capability to automatically power-on and power-off based on the VAD interrupts. As an example, the TAx511x or TAx521x system can be set up to monitor VAD activity on a digital microphone channel and then power on the analog microphone channels based on the VAD trigger. By using the VAD to control the ADC power-on and power-off behavior, power savings can be achieved, compared to leaving the ADC powered on at all times.

The VAD has three modes that determine the behavior of the ADC record path:

  1. Auto Mode: Voice triggered VAD interrupt based ADC power on and power off. The ADC record path is active only during voice activity.
  2. User Mode: Voice triggered VAD interrupts have to be monitored by the host and the ADC record path is powered on or off through I2C commands.
  3. Intermediate Mode: Voice triggered VAD interrupt powers on the ADC record path and the ADC continues recording until powered off through host I2C commands.

Note that in all modes, the device generates an interrupt on the configured pin which can be sent to an external DSP or SOC.

The salient features of VAD are as follows:

  • Clock configuration for VAD includes:
    • VAD processing using internal oscillator (Target mode)
    • VAD processing using external clock on BCLK input (Target mode)
    • VAD processing using external clock on MCLK input (Controller Mode)
  • In external clock configuration, the VAD processing clock frequency can be adjusted to meet the system power demands.

  • Automatic switching between VAD mode and Record mode - The system switches from VAD mode to record mode upon voice activity and switches back to VAD mode upon no voice activity in Auto VAD mode.