I have been tasked with redacting sensitive information from some audio files. The files are of mixed format (WAV, MP3) etc. Once I open the file in Audacity, I am selecting the part to be redacted, then select Generate–>Tone. I leave the default as sine and select 1000 Hz as the frequency. and click ‘ok’ .
Audacity shows in the waveform the 1000 Hz waveform where the selected audio was. I then exported the file back to it’s original format type.
Does anyone know if the redaction could then be reversed from the exported audio? For example, is there a way to ‘unmask’ the audio that was redacted?
That method does not “cover up” the original audio, it “replaces” the original audio. The original audio will be completely deleted and have a 1000 Hz sine wave in its place. After exporting as a normal audio file, there will be no way to reveal the original audio because it is no longer there.
I then exported the file back to it’s original format type.
Hopefully with a different filename. Open and play the corrected file after you export it. I would do it in a different player like Windows Media or iTunes. Nothing like sending the client the wrong version of the file.
“Wait, didn’t you listen to it before it went out?”
I think it is possible to scramble the audio samples in such a way that the words are unrecognisable and unscramble them with a plugin that reverses the process. I can vaguely recall a technique posted somewhere for doing that but can’t find it right now.
The most obvious (but easily defeated) way is to Effect > Reverse the audio, then the recipient would Effect > Reverse the file you sent them to make the words intelligible again.
The simple answer however is to export a WAV of the uncensored audio before you begin.