python pydub generate audio file from keyboard presses

Question

I've written a program to detect my keyboard presses and generate an audio file with pydub from this recording. When I inspect my two audio files (generated by python and recorded by smartphone) in audicity it's quiet obvious that the recording from my smartphone has spikes (higher amplitude) and some sort of fade in and fade out on every longer "typing sequence", but my generated file has not. Is it possible to detect and adjust the amplitude of a "typing sequence" in pydub, that it sounds more natural?

Here is my code to detect my key strokes and to generate my audio track:

def getTimestamp():
  return round(time.time()*1000)

if __name__ == "__main__":
  recording = {}
  start, stop = (0, 0)
  running = False
  while True:
    if keyboard.is_pressed("f10"):
      start = getTimestamp()
      print("Recording...")
      running = True
    if keyboard.is_pressed("f12"):
      stop = getTimestamp()
      print("Stopped recording!")
      break
    if running:
      current_time = getTimestamp() - start
      recording[current_time] = keyboard.read_key()
  # create audio file
  createSound("./test.wav", start, stop, recording)
  print("Done!")

The createSound() function looks like this:

from pydub import AudioSegment

def createSound(filename, start, stop, recording):
  # create empty track
  track = AudioSegment.silent(duration=(stop-start))
  for timestamp, key in recording.items():
    keypress = AudioSegment.from_file("./click.wav")
    # add keypress to track
    track = track.overlay(keypress, position=int(timestamp))
  # export track
  track.export(filename, format="wav")