PZM-10 impulse response for convolution

1 reply to this topic

#1 bobonofx

  • Members
  • 1 posts

Posted 07 April 2008 - 06:26 AM


I would like to know if someone has already measured the impulse response of the PZM-10 microphone.

If yes, can you send it to me?

I want to use it for convolution

Best regards


#2 brubart


    Power User

  • Members
  • PipPipPipPipPip
  • 52 posts

Posted 07 April 2008 - 09:22 AM

It sounds like you want a wave file recording of the PZM-10 responding to an impulsive sound source like a gunshot or a balloon popping. If you use convolving software to convolve (multiply) that IR with a speech wave file, you will get the sound of the speech as picked up by the PZM-10.

Sorry, we donít have that data; but you can get very nearly the same result by equalizing the speech according to the frequency response of the PZM-10 that is published in its datasheet.

You would need to record a speech wave file anechoically (or nearly so) with a flat-response omnidirectional measurement microphone. Then equalize that recording with a response curve that matches that of the PZM-10. The resulting recording would be the sound of speech as reproduced by the PZM-10. That would be an anechoic recording, which you would then convolve with the impulse response of the room in which the speech is going to be. The end result would sound like speech in that room as picked up by the PZM-10.

When you record speech with the measurement microphone, the orientation of the talker to the mic should be the same as it would be in the real-life situation. For example, if the PZM-10 will be 45 degrees off the axis of the talker's mouth, the measurement microphone should be too, in order to capture the off-axis spectrum of the voice.