Hello, I have a recording of a speech that’s not very clear. Running the spectrum analysis I get this result, with all the audio included between 0 and 4000 Hz. Is there a way (if it makes sense to get a more clear sound) to stretch this range extending it to the highest frequencies, like one does in Photoshop using the Curves or Levels tool? I experimented a little with Equalizer but didn’t get a good enough result. Thank you very much.
I forgot: I’m using Audacity 2.0.5 on Linux Mint 17 XFCE, downloaded from the standard repository.
P.S.: please forgive me if I did any language mistake, I’m not an English native speaker.
P.P.S.: how can I insert my signature in this forum? I couldn’t find the dedicated text field.
The “clarity” of speech recording is in frequencies above 4000 Hz, but as shown in that spectrum, frequencies above 4000 Hz are completely absent.
If the frequencies above 4000 Hz were just “a bit low”, then you could boost them using the Equalization effect, but you cannot boost something that is completely missing. So the short answer is no, I doubt that significant improvement can be made.
Too many people were abusing the “signature” field by using for spamming, so reluctantly we had to disable it except for long-standing forum users.
Thank you very much for the very fast answers. Just out of curiosity: so doesn’t even exist an audio filter (not only in Audacity) that expands the frequency range like in this picture I created? And if it should exists, what would be the result when listening to the processed audio? Thank you very much again.
Uh, by the way, I asked the question about the signature because I liked to insert the “not English” thing in all of my posts.
There are microphone systems and recorders that will not produce crisp, clear sound. I have a portable recorder that has the digital sample rate baked in. The recording quality will never get better than FM Radio and it doesn’t matter how I filter or process it.
You should change how you record the work. Telephone quality is 3000 and AM radio is 5000, so your work is between those two.
Picture inserted in-line using the “Img” tag button:
It is possible to “expand” the frequency range upward, but not by filtering, and it will probably not produce a pleasant or useful result.
Try applying the “Change Pitch” effect (Audacity Manual) using a setting of 12 semitones, then look at the spectrum.
Ok, I got it… Thank you very much Steve.
@kozikowski: you are right, the problem is that the person who recorded the speech was not me… Thank you again, bye.