It is possible somehow, e.g. Setting a noise floor with VST? So I don't want a noise gate so that nothing or everything can be heard, but that sounds below XdB are cut out. So it would have the same effect as if I replaced the microphone with a microphone with a lower sensitivity. So when I speak loudly, it would not let everything through, including the noise, but it would also cut out the soft sounds, that is, the noise. Why can't this be done? Or is there such a thing, I just don't know about it? Yes, and it should be live. But one of these would be good for editing as well.
I thought the old Phase Linear "Autocorrelator" did this. Split the audio into frequency bands, where each band had a threshold, anything below which only silence was output. I had one 30 or so years ago. You can still get them on ebay https://www.ebay.com/itm/354535953143
I wouldnt know how to do it in software as a VST, but I'd bet the guy who makes the Reaper stuff could.
I wouldnt know how to do it in software as a VST, but I'd bet the guy who makes the Reaper stuff could.
Expander. It's the name of the kind of treatment you are looking for.
Help to perform same task as a compressor but instead of taking care of loudest signal it does it the other way around in taking care of softest. Most dynamic treatments belong more or less to gate familly so it often offer gate or comp facility too.
There is other way to do it ( read it can be complex to implement...but fun to do too!) but try this first. Could you describe more clearly the reason you need it and domain of use ( radio, music recording,...).
Junger audio had a unit ( hardware) which was almost mandatory for broadcast: seen it either for tv/radio/mastering. Can't remember the exact ref but it shouldn't be difficult to find. Maybe b or j 40... something like that. It had been cloned in vst.
Edit: Yeah Flux audio did the vst version of it.
Some reading:
https://www.google.com/url?sa=t&sou...IQFnoECBYQAQ&usg=AOvVaw3qTrYAehotlNkonxOW7eK4
Help to perform same task as a compressor but instead of taking care of loudest signal it does it the other way around in taking care of softest. Most dynamic treatments belong more or less to gate familly so it often offer gate or comp facility too.
There is other way to do it ( read it can be complex to implement...but fun to do too!) but try this first. Could you describe more clearly the reason you need it and domain of use ( radio, music recording,...).
Junger audio had a unit ( hardware) which was almost mandatory for broadcast: seen it either for tv/radio/mastering. Can't remember the exact ref but it shouldn't be difficult to find. Maybe b or j 40... something like that. It had been cloned in vst.
Edit: Yeah Flux audio did the vst version of it.
Some reading:
https://www.google.com/url?sa=t&sou...IQFnoECBYQAQ&usg=AOvVaw3qTrYAehotlNkonxOW7eK4
Last edited:
Editing?
Isn't what automation was designed for? 😉
Dynamic treatments are nothing more than automated faders reacting followig a pre set of instructions.
You can do excatly the same by automating a fader during mixdown.
All daw editing software include a way or another to do this.
Please describe exactly your use and the issue you face as well as gear and soft.
Isn't what automation was designed for? 😉
Dynamic treatments are nothing more than automated faders reacting followig a pre set of instructions.
You can do excatly the same by automating a fader during mixdown.
All daw editing software include a way or another to do this.
Please describe exactly your use and the issue you face as well as gear and soft.
I wrote it for both editing and live, because it would be good for both. For many years I have tried e.g. to denoise a cassette, and it would have been very useful to denoise a noisy file, but I could only use a noise gate and a noise filter, which I don't like, because I don't like the noise gate because only the pause is not noisy, what I would listen to is just as noisy. The noise filter also tends to reduce the sound quality. And the noise filter can only be used for microphones and live, or only for editing. And in most cases I just wanted to remove the low noise. The other reason why I highlighted that e.g. noise filters, in addition to making the sound worse, can often only be used for files, not live. But now my specific goal is that if there is a sensitive condenser microphone, then it should be good for everything! So, if it was in space, without background noise, it would still be great, but if it is in a plain room, with a PC, a bare wall, etc., then it would be possible to extract sound from it as if it were a dynamic microphone. You don't have to buy a dynamic microphone just because it turns out that it hears less noise. Because in my opinion, what it does because of the physical properties of the dynamic microphone is that it makes your ears worse, and what you can't hear, you can't amplify, so it doesn't hum with the same loud speech. I don't know how sudden the dynamic microphone's hearing loss is, but it should be similar. However, I just looked up the Expander and it is quite similar to the noise gate. This is not what I want. I don't like it being silent and then the sound is noisy. I want to take the noise out of the noisy voice. Not just from breaks. The sound contains the quieter and louder signals at the same time. I want to reduce or delete the quieter signals. In the whole. The whole thing has that level of background noise throughout. I want to remove it from the whole thing, not just from the break. I drew it. I want to remove sounds with volume below the red wave from the sum of the waves.
Hi,
Interesting pov i don't share on some points... i have to bring kids to school and will answer to you after it's done.
Interesting pov i don't share on some points... i have to bring kids to school and will answer to you after it's done.
Ok. For me if a expander doesn't works for you then... i can tell you don't know how to set it up.
There is a parameter called 'range' which give you hands on the maximum attenuation the process will give. As i told those are used for broadcast on the 'on air' processing chain ( tv, radio,...) total cuts ( gate behaviour) is forbidden and it is defined by this 'range' parameter... at most it'll attenuate the sound below treshold to limit of audibility ( by masking effect thanks to background music or noise).
The point is not to 'filter' the unwanted sound but to 'push it' well behind the other sound source a program have and so 'mask' it.
You talk about 'noise filter' but please give exact ref to the plug in you use as there is a wide range of tools and all doesn't behave the same ( or have same target and working principle).
Low noise? Low in level (amplitude) or frequency ( low end of frequency spectre)?
So now we are talking 'restoration' tools and yes, if you don't know how to apply them then result can be worse than initially.
I agree real time 'noise removal' is not easily accessible for amateur ( there is exception in pro world with things like Cedar but it is way out because of budget involved...
It's a fantasy. Condensor mic have advantage over dynamic mics but that won't say it's better overall. Both have pro and cons but the essential point is how they are adapted to your source ( eg voice) and the use you make of it ( speech on radio or tv, a singer, etc,etc, all have different requirements).
To give an easy analogy saying a Formula1 is a better vehicle than a truck without giving a use for it have no sense ( try to go at grocery store with an F1...).
OK. So you have already given half your own answer: treat the issue at source. If you have a noisy pc and bare walls then your mic is going to display that...
First try to make computer and others source of noise quiet ( by changing, fans, psu or define a 'cooling profile' ( if fans are pwm controled) to lower noise floor, rearrange inside the room, etc,etc,...).
If there is acoustic issue with your room then until they are treated by acoustic means it'll be difficult to treat afterward ( almost impossible).
It's too broad a subject to give you answer here but both advice are to be tryied before anything else in my view.
Your opinion is false on this. We use dynamic, ribbon, condensor for other means that the one you thought about.
Each family as it's pro and cons and they have to be adapted to the situation...
Dynamic mics are perfectly adapted to voice. I recorded whole album with SM7 eg. It wasn't classical music ( hip hop/rap) but it gave the opportunity to singer to feel good and nice ( he could move over the recording studio handling the mic, makes him more confortable than staying without moving behind a condenser mic = better takes in the end).
Sure on the C800g his voice sounded 'technically' better but as was not confortable during takes they were unusable... and in the end you won't hear this much a difference once into context of track
There is no hearing loss as you said but physic dictate some points: moving mass is heavier for dynamic mic so they don't go as high as condenser and can be a bit less sensitive. But don't expect as much differences as the numbers could give on a technical spec sheet: once level compensated ( both mic to same playback level) nothing tells you you'll choose a condenser in blind test ( and believe it or not we usually do this kind of things before recording a singer: set up a number of mic and compare them blind until finding the one which 'works' with the singer's voice).
So you don't have tried one ( see previous answer about 'range' parameter).
So you are looking to something else which bring us back to 'restoration' tools.
When i leaved audio industry the standard software for it was Izotope's RX. It bet it still is. In sound for movies it must be a full time job as RX operator.
It's not cheap, not easy to set up but it's the more powerful software for this kind of treatments.
In my experience it doesn't work this well real time* but offline it makes wonder if you know how to use it. The drawback: it takes time.
It have a steep learning curve and the nature of treatments can't be made in a hurry if you want 'stealth' treatments.
What you drew and ask are known as 'demixing' tools ( being able to separate sounds on a mixed track: drums from singer, guitars,etc,etc,...). I know there is research on this and been lucky to see what IRCAM had ten years ago for it. They had impressive results but still not to the point it 'worked'. As far as i know it's still a fantasy ( despite some polyphonics treatments like Melodyne and others...).
Maybe it'll happen in the next years though...
Gyulank, you still haven't given a clue about which context you need these treatments: radio ( streaming, podcast,...) music, movie,...
It might not seems important to you but it is for answer.
* for some kind of non changing repetitive sound profile it can work. Like AC or some regular motor based sound source. As soon there is variations in the rpm of sound source it's much less easy to set up...
Last edited:
I think we're talking past each other. Maybe because of google translate. Sometimes I corrected it though. I would highlight 2 separate things and describe them in more detail.
1.; All other parameters are irrelevant from the point of view of the question. In the comparison, assume a straight frequency response for both microphones! The difference should be that the sensitivity of the dynamic one is lower, the line is lower! Or if it's not straight, let's say you like the sound of both! Let's not get stuck on that! The frequency response is not necessarily straight, but there are certainly many condensers and many dynamic microphones with similar +/- deviations, so we can safely ignore this difference, plus it exists between dynamic-dynamic and condenser-condenser as well. The fundamental difference between dynamic and condenser is that dynamic is more insensitive and the noise floor is higher.
2.; Let's take as a basis a smooth sound that can be heard anywhere! It contains all kinds of frequencies and volumes. This is a wave of various frequencies, amplitudes, and phases, which is summed up on the membrane in a jumbled wave, which contains all the audible sounds. In analog technology, every frequency has a volume at every moment, and in digital technology, too. So there are soft and loud sounds at the same time. The EQ e.g. they increase or decrease the amplitude of waves of certain wavelengths. The noise filter does the same, set to noise. With analog, I even understand that it is easy to make frequency filters and amplify or reduce them. Now suddenly I can't think of an analog solution for what I want. That doesn't mean it's impossible. But digitally, I don't see any obstacles. All sounds are recorded (also live). Encoded in PCM, just like everything else, the volume levels could be selected and modified. There are specific values here, each frequency has a volume level. You just have to choose the right domain. And then you could do anything with it in the same way, delete it, turn it down, reduce it with different slopes, even in a frequency-dependent way.
And so in the sound of the condenser microphone, the extra noise would be useless, because the noise floor could be adjusted and it would cut it out.
1.; All other parameters are irrelevant from the point of view of the question. In the comparison, assume a straight frequency response for both microphones! The difference should be that the sensitivity of the dynamic one is lower, the line is lower! Or if it's not straight, let's say you like the sound of both! Let's not get stuck on that! The frequency response is not necessarily straight, but there are certainly many condensers and many dynamic microphones with similar +/- deviations, so we can safely ignore this difference, plus it exists between dynamic-dynamic and condenser-condenser as well. The fundamental difference between dynamic and condenser is that dynamic is more insensitive and the noise floor is higher.
2.; Let's take as a basis a smooth sound that can be heard anywhere! It contains all kinds of frequencies and volumes. This is a wave of various frequencies, amplitudes, and phases, which is summed up on the membrane in a jumbled wave, which contains all the audible sounds. In analog technology, every frequency has a volume at every moment, and in digital technology, too. So there are soft and loud sounds at the same time. The EQ e.g. they increase or decrease the amplitude of waves of certain wavelengths. The noise filter does the same, set to noise. With analog, I even understand that it is easy to make frequency filters and amplify or reduce them. Now suddenly I can't think of an analog solution for what I want. That doesn't mean it's impossible. But digitally, I don't see any obstacles. All sounds are recorded (also live). Encoded in PCM, just like everything else, the volume levels could be selected and modified. There are specific values here, each frequency has a volume level. You just have to choose the right domain. And then you could do anything with it in the same way, delete it, turn it down, reduce it with different slopes, even in a frequency-dependent way.
And so in the sound of the condenser microphone, the extra noise would be useless, because the noise floor could be adjusted and it would cut it out.
Last edited: