Algorithm 1 Voice Activity Detection (VAD)

$input\leftarrow$ audio file $rms\leftarrow$ frame energy($input$) $threshold\leftarrow 0.05\times max\left(rms\right)$ fori in $range\left(len\right(rms\left)\right)$ do    if $rms\left[i\right]\ge threshold$ then      frame i stored in $speech$    else      frame i stored in $nonspeech$    end if end forreturn$speech,nonspeech$