Hello there! this is my first question here and I would really appreciate any help
I am devoloping an application in C for removing the noise from a batch of files located in an specific folder, I would like to script the Audacity's Noise Removal feature since it's very effective. I have done some reading about the mod-script-pipe, but I think I've read that this feature can't be scripted, So I was wating if someone could redirect me to the dsp source code (If I am able to use it) or give me some ideas of how can I script Audacity's noise removal.
Thanks in advance!
Scripting Noise Removal
Forum rules
If you require help using Audacity, please post on the forum board relevant to your operating system:
Windows
Mac OS X
GNU/Linux and Unix-like
If you require help using Audacity, please post on the forum board relevant to your operating system:
Windows
Mac OS X
GNU/Linux and Unix-like
Re: Scripting Noise Removal
It's here: http://code.google.com/p/audacity/sourc ... %2Feffectsjchayan wrote:if someone could redirect me to the dsp source code (If I am able to use it)
You can use it subject to the terms of the GPL v2 license.
The effectiveness of the Audacity Noise Removal effect is highly dependent on the "Noise Profile" matching the actual noise that is to be removed.
You might find it easier and more convenient to use the Noise Removal effect that is available in the command line program SoX.
9/10 questions are answered in the FREQUENTLY ASKED QUESTIONS (FAQ)
Re: Scripting Noise Removal
Having similar issue as jchayan I decided to give a go to 'SoX', but regrettably in comparison to Audacity it has quite poor performance. It's noise reduction method is no better than my simple, home-cooked in Python algorithm for spectral subtraction. In fact, since results are the same, I suppose it is the same technique underneath.
I suppose one of keys to success is, as quoted from Audacity web site: "During the noise removal phase, we start by setting a gain control for each frequency band such that if the sound has exceeded the previously-determined threshold, the gain is set to 0, otherwise the gain is set lower (e.g. -18 dB), to suppress the noise. Then frequency-smoothing is applied so that a single frequency is never suppressed or boosted in isolation, followed by time-smoothing so that the gain for each frequency band moves slowly."
Do you know of any other tool, preferably one that has Python wrappers, that deals with noise reduction in a similar manner to Audacity and allows scripting?
Thanks!
I suppose one of keys to success is, as quoted from Audacity web site: "During the noise removal phase, we start by setting a gain control for each frequency band such that if the sound has exceeded the previously-determined threshold, the gain is set to 0, otherwise the gain is set lower (e.g. -18 dB), to suppress the noise. Then frequency-smoothing is applied so that a single frequency is never suppressed or boosted in isolation, followed by time-smoothing so that the gain for each frequency band moves slowly."
Do you know of any other tool, preferably one that has Python wrappers, that deals with noise reduction in a similar manner to Audacity and allows scripting?
Thanks!