Taperssection.com
Gear / Technical Help => Post-Processing, Computer / Streaming / Internet Devices & Related Activity => Topic started by: moondust.and.solitude on October 17, 2020, 10:40:45 PM
-
(see picture)
I'm working on transferring an old cassette recording of an FM radio broadcast. It sounds great except, for a lack of better wording, a "shrill" artifact. I'm not the best at EQ and looking for some advice. This is probably a simple fix, but I'm not able to dial it in yet. Looking for some sound advice (nice play on words there) :bigsmile:
-
Others will probably have more expert advice, but in Audacity I usually just highlight the offending peak and shrink it using Amplify (i.e. Amplify to a lower level) to bring it down to the level of the rest of the track. Have used this for some nasty vocals.
-
reduce treble above 15,750 with a low pass filter?
FM radio only goes to 16kHz anyhow, so that looks like it could be modulation nonsense
-
Looks like the pilot tone of 15.7khz from ntsc TV.
Just do a sharp EQ cut in a parametric eq, or notch filter.
-
Looks like the pilot tone of 15.7khz from ntsc TV.
Just do a sharp EQ cut in a parametric eq, or notch filter.
steeper filter curve = more phase shift
#NobodyCares
-
If we do not mix two close microphones, then the phase shift from eq should not matter. Or am I wrong?
-
If we do not mix two close microphones, then the phase shift from eq should not matter. Or am I wrong?
Do your speakers have a crossover network to divide woofer business from mids and or tweeters? That's also a mix, so phase delay can cause mismatch there.
-
Thanks Morst. I don't know crossovers. Probably these crossovers are also adding phase shift to the signal. But eq phase shift alone don't cause comb filtering. The comb filtering would be created if we mixed the original signal and the shifted signal together.
-
Thanks Morst. I don't know crossovers. Probably these crossovers are also adding phase shift to the signal. But eq phase shift alone don't cause comb filtering. The comb filtering would be created if we mixed the original signal and the shifted signal together.
A crossover is how one signal divides in order to drive a speaker cabinet which contains a woofer and tweeter, or woofer-midrange-tweeter, for example.
A simple crossover circuit would roll off (remove) the low frequencies from the tweeter so it only gets high frequencies, and would roll off the highs from the woofer, allowing it to pass the low part of the signal.
The signals going to each driver are non-identical, and mix acoustically in the room. Thus, what the listener hears is in fact a mix of the sound coming from each driver.
Even a simple two-way crossover will cause different frequencies to arrive at different times.
Intuitively, I desire to create the least time smear possible, thus I eschew excessive equalization, or other processing which might result in [frequency] group delay.
(http://www.learningaboutelectronics.com/images/Speaker-crossover-network-circuit.png)
http://www.learningaboutelectronics.com/Articles/Speaker-crossover-network-circuit.php
-
Thank you Morst for explanation and schematic. Perhaps I understand what you mean. Dividing the signal into two drivers causes problems in the frequency range where the two drivers overlap. Comb filtering can occure in this frequency range in the room. And it is possible to hear it.
But EQ only works with one signal. There cannot occur comb filtering. I don't know if anyone can hear this little phase shift caused by the equalizer. I can't even think of a way to test it. Maybe by listening to standard eq and linear eq. But linear eq brings some artefacts into the signal. The comparison would be inaccurate.
-
Thank you Morst for explanation and schematic. Perhaps I understand what you mean. Dividing the signal into two drivers causes problems in the frequency range where the two drivers overlap. Comb filtering can occure in this frequency range in the room. And it is possible to hear it.
But EQ only works with one signal. There cannot occur comb filtering. I don't know if anyone can hear this little phase shift caused by the equalizer. I can't even think of a way to test it. Maybe by listening to standard eq and linear eq. But linear eq brings some artefacts into the signal. The comparison would be inaccurate.
Interesting points about comb filtering.
I'm not primarily looking at it from the perspective of comb filtering, I'm just trying to simplify the ideal signal path to result in less distortion, including phase distortion or as I said, group delay
I recently (the past 5 years) started learning the science of Meyer SIM and Rational Acoustics SMAART analysis systems, and one goal of system optimization seems to be flattening out phase artifacts at the perspective of the listener.
Intuitively to me, providing a signal to an optimized system which has been time smeared is not as ideal as providing one which has not.
I hope this is informative and not overly confusing. The stuff I'm learning on the subject is somewhat mind blowing, so I'm sure I don't understand it all, and I love discussing it because it helps me to understand more.
Here's a good intro video with Merlijn Van Veen of Meyer Sound, posted in may 2020.
https://youtu.be/cfhqHxod9D0 (https://youtu.be/cfhqHxod9D0)
-
I suspect one of the reasons I don't like high-pass filtering as standard procedure and use it only when necessary, even when listening on systems that don't really have deep bass extension, is the repercussions of excessive group delay imposed by the filtering. But that's little more than a hunch. Would love to really know.
I've a few recordings where the bass behaved really oddly. In general they were outdoors with big PAs. Plenty of apparent level yet sounds odd, weak, and mismatched to the rest of content. I think those have to do with my recording accurately capturing a particularly bad subwoofer alignment of the PA. But again.. little more than a guess.
[edit- thanks for the Meyer link, Morst, looking forward to checking that out]
-
Thanks to Morst and Gutbucket. I see I have joined the discussion that is not as simple as it seemed to me in the beginning. These are very interesting things. I'll watch this video.
-
Thanks to Morst and Gutbucket. I see I have joined the discussion that is not as simple as it seemed to me in the beginning. These are very interesting things. I'll watch this video.
Honestly, even the intro video might even be too advanced and nebulous to really apply to the topic of this thread, but I wanted others to have access to the same info that I do. I would love to hear comments and try to answer questions about it... But since this is an EQ thread, let's discuss the Meyer videos in a thread I made about them:
https://taperssection.com/index.php?topic=195145.0