Software to Apply FIR Filter - No Math

Altera Quartus includes a Megawizard FIR compiler for free.

Andy

Apply how? Do you want to listen to an audio file with & without? Do you want to design FIR filters for DSP on some sort of chip? If so, what sort of chip? If on a PC, do you want to do the filtering in real time (i.e. do you want it streamed), or can you do it as a batch process on a file?

I'm taking the liberty of cross-posting this response to comp.dsp, because I think you'll get a lot of answers there. But you might want to describe at the top level what you're actually doing, and what you want to get out of it (i.e., "I want my dog to bark into the microphone while I hear the filtered bark in my headphones" or "I want to low-pass filter 'Smoke On The Water' to see what it sounds like without the high frequencies", etc.).

What high frequencies? Those were gone about 5 years ago...

and they never existed in that heavy rock (like Deep Purple) before the onset of the screechy heavy metal from the likes of Van Halen, Megadeath, etc...

I have the opposite; there are some tones that are there all the time now.

Eric Jacobsen Anchor Hill Communications

Maybe VST plugins and a free audio editor might be what you're looking for.

Yes, I would like to use a Windows PC sound card input, and then FIR process the acquired signal in some yet-to-be-determined software.

I am looking for a GUI based FIR program that will allow me to filter out various frequencies and observe the effect in real time, both audibly and in spectrum analysis software.

There is just not enough time to learn everything, like FIR coding, so hopefully there is an existing solution out there.

Ken Rockwell

I know there are, and I've seen them discussed on comp.dsp -- which is why I went and cross-posted you into there in the first place.

While we're waiting for someone more knowledgeable about that sort of thing to reply, try Googling on "audio workshop", "dsp workshop", "dsp audio workshop", etc.

Note that for audio, classical symmetric FIR filters do not necessarily sound good -- they have a "pre-echo" effect that comes from their symmetry around some delay point, and this pre-echo does not sound good. If your primary goal is voice (or other audio) processing for communications you may not care. If it's music, then you may want to use asymmetric FIR filters or IIR filters.

Don't know about FIR, but there are programs that will take a sound card input and filter it. NOt free, but cooledit has a filter function that lets you drag the filter curve around. Just tried it on an open file. Not sure how to make it run in real time. Maybe there's a similar free alternative. You'll have to google graphic equalizer.

Tim,

I've heard that said in the past, but I'm not sure I believe it. I mean, of course you have a "preresponse", but is it audible?. Do you have a pointer to a sound clip that demonstrates this?

Audacity

Does it really have to be real time. Can't you record some audio then try filtering that file? If so, just load audacity. There is a windows version.

If you want to FIR filter an audio stream, you can do that with sox. The filtering is real time, but you need to design the FIR tap coefficients, and you can't do that on the fly the way sox is set up.

Do a search for "fir " on this page.

Sox is standard on linux. You can load it for free for windows. I've used it on both platforms. It is an incredibly versatile program if you are the type that isn't afraid of command lines.

Designing FIR filters isn't that hard. With the Remez Exchange algorithm, you can design filters you don't like all day. Matlab has this built in. Octave as well.

Searching the internet, I found this website that offers free FIR filter design software:

I'm not particularly surprised this exists as free software. Park McClellan (Remez) is in the Oppenheim/Shafer DSP book.

Do the filters work in real time? I didn't think so. Same for CoolEdit/Audition.

Ken Rockwell

Yes, that was the crux of my inquiry. And to do it in a GUI without having to code something from scratch.

It always amazes me in electronics that some of the most obvious time-saving things never seem to get develped to the point where a non-specialist can just pick it up and use it.

Ken Rockwell

So, in other words, you expect someone more qualified than you to spend his expensive time so that you can save some of yours? Get real.

Jeroen Belleman

Can't speak for him, but that's what I expect. That's what freeware is all about. Virtually everything I do on my computer is the result of someone else spending his valuable time so I don't have to.

And my interpretation of the request is to find a pointer to something already done.

On a sunny day (Tue, 04 Feb 2014 18:49:28 +1100) it happened snipped-for-privacy@imagenet.com wrote in :

Because a 'non specialist' would use it the wrong way.

Yes. Like that time I explained some physicists that to equalize the frequency response of two channels of some measurement instrument, they'd just have to filter the digitized data stream of one of them.

They chose to *FFT* the data, apply the filter's magnitude response in the frequency domain, and then FFT it back to time domain. The way they did this totally messed up the phase response and made the data entirely unusable, not to mention the considerable time spent doing the transforms.

My solution was an IIR filter in four lines of C-code that did the right thing in less time than needed to display the results. Oh well.

Jeroen Belleman

On a sunny day (Tue, 04 Feb 2014 10:06:29 +0100) it happened Jeroen Belleman wrote in :

Yes I have tried that with video, it worked, but there were some artifacts: