Is it possible to make instruments sound natural through audio editing like this?

[Paul R Johnson]

Right - here is me playing (badly, because I had 5 mins) flute, as quetly as I could, clarinet low, and a treble recorder, not the usual descant. Can you hear the difference? Random noodling and reverb - that's it.

[Ryan Elder]

Oh no audio link was posted, if you meant to post one?

[Paul R Johnson]

Yes sorry - Every time I tried, my browser crashed - I'm back on the MacBook - try here.
www.eastanglianradio.com/ryan1.mp3

[Ryan Elder]

Oh all good thank you very much for posting that for me! So the recorder starts at about 0:24 then? yes that recorder sounds different compared to the ones I have been listening to. What kind of recorder is that or what is the name of that type?

[Brian Drysdale]

The clue is the name "bass" recorder, which the recorder you've been listening to.

[Ryan Elder]

Oh okay. The YouTube videos I linked before have bass recorders but they sound more clarinet like for some reason. Is it because they're being played in the lower notes?

Were you playing the higher notes more so?

[Brian Drysdale]

Here's video on the clarinet

https://philharmonia.co.uk/resources...ents/clarinet/

[Ryan Elder]

Oh thank you. Well maybe the bass recorder is not the right sound I am looking for perhaps. I know I want a low woodwind sound, that you would get from blowing air through a woodwind. But maybe the bass recorder would work in the context of the music I have in mind, if it was being played in that fashion.

[Paul R Johnson]

The one I played was in it's lower range - and it's a treble (sometimes called alto) There's a picture here of treble, tenor and bass - the bass looking like a bassoon, with a crook.

https://usercontent.village.co/b0a8d...fa87-image.jpg

Bass recorders really do not sound like reed instruments. The important thing is that in one of your clips, it wasn't a real instrument at all, but a synthesised sound, but does it matter? I really don't think so. I have three saxophones - soprano, alto and tenor. I'd like a baritone but can't afford one. They all sound different to me, but sometimes in movies you will see somebody playing a tenor, but hear an alto. Car experts notice when a car in a movie is a V8, but we hear a V6. My friend the aviation buff can spot the wrong engines in war films.

When we have sound for movies, the critical thing is it fits. If it's supposed to be haunting - if it is, it worked. Often sounds are contrived and artificial. I really don't think that telling a composer that you have a cello in your head works, if your idea of a cello is different from mine. You can't hear the difference between flute and reed. That's fine, lots of people can't, but that means your judgement is a bit flawed. A bit like a colour blind person picking fabrics and paint colours.

Do you know anyone who has a bass recorder? I don't?

[Ryan Elder]

Oh okay thanks. Yeah I am not an expert on the categories of instruments such as recorder, reed instruments, etc. I was just picking based on sounds I felt would go with the tone and feeling of what I wanted, while played in that fashion of course.

I actually thought that the bass native American flute was a synthesizer at first, but the composer told me it was a bass native American flute. But if he is incorrect and it's a synthesizer, than we can use that then.

However, since I want the flute to go into overblowing in certain parts. Can a synthesizer, synthesize overblowing? I'm assuming it can't because you never hear a synthesizer overblowing in any musical pieces that I am aware of, unless there are any?

[Paul R Johnson]

Of course they can! Well, to be fair, some can if they are designed to do it, but you are still confused. Synthesis or sampling is now often a bit of both - and the samples are edited so maybe you hear the first bit of sample 20394 crossfaded into 21928, and then when your hand comes off the key, it ends with the 25341 full stop!

Who actually cares? You are trying to categorise things. You have no idea if you have even heard a synthesised sound, and not noticed. Maybe try to treat it like Foley - you see the bird, you hear the wings flap. The flaps might well be somebody in the Foley studio with old fashioned leather motorcycle gloves. It doesn't matter.

Here's a project of mine that originally had images - this is the soundtrack, and the voice is a couple of hundred phrases joined up, with other parts synthesised. Is it real? No idea - sounds real.
Then there is the second one - try this one, real, sampled or synthesised? Synthesis does not have to sound like Jean Michel Jarre

YouTube (Short URL)

YouTube (Short URL)

[Ryan Elder]

Originally Posted by Paul R Johnson

Of course they can! Well, to be fair, some can if they are designed to do it, but you are still confused. Synthesis or sampling is now often a bit of both - and the samples are edited so maybe you hear the first bit of sample 20394 crossfaded into 21928, and then when your hand comes off the key, it ends with the 25341 full stop!

Who actually cares? You are trying to categorise things. You have no idea if you have even heard a synthesised sound, and not noticed. Maybe try to treat it like Foley - you see the bird, you hear the wings flap. The flaps might well be somebody in the Foley studio with old fashioned leather motorcycle gloves. It doesn't matter.

Here's a project of mine that originally had images - this is the soundtrack, and the voice is a couple of hundred phrases joined up, with other parts synthesised. Is it real? No idea - sounds real.
Then there is the second one - try this one, real, sampled or synthesised? Synthesis does not have to sound like Jean Michel Jarre

https://youtu.be/SdwHQymp6AU


https://youtu.be/mfbpTmBkwRU

Oh okay thanks. Well I would guess that the first one is real, and the second one is sampled, is that right?

When you said that one of the clips I provided, the instrument was actually synthesized, do you mean the flute at 0:38 in this clip:

The Replacement Killers soundtrack part 3 - YouTube

When you say synthesized, do you mean it's sampled, or done on a synth completely?

[Paul R Johnson]

You are STILL missing the entire point - you clearly don;lt understand synthesis vs sampling, and the fact that hardware synthesisers are largely gathering dust as they are inside computers, alongside samplers. The sound at 38 seconds is layered, effects treated and evolutionary - so probably generated as programming number XXX in ZZZ instrument inside a computer. It could be a prophet from the 80s on a stock patch, or a if recorded today, one of a thousand different sound sources. It does not matter how it was generated, but it certainly doesn't sound like a real instrument with a microphone, that's for certain.In those clips, the classical piano was synthesised NO samples used at all in it's construction - it's called a virtually modelled sound - as in, it's synthesised. The less realistic flute is actually a sample. The voices in the other were real voices, but the singers sang every note, and every phrase in ways that could be chopped up and reconstructed, but there is another similar one where the words are synthesised, and I cannot tell the difference.

Your replacement killers track was produced in 98, so almost certainly synth based, because samplers were not that advanced back then.

[Ryan Elder]

Oh okay. I thought that the difference between synthesis and sampling, is that synthesis is made from a synthesizer, and sampling is recordings of the notes of the real instrument. That is what the composer said anyway, but is that wrong, if I don't understand?

[Paul R Johnson]

Synthesis is the creation of a sound from (in an analogue device) combining different waveforms. The synthesiser can be hardware with oscillators, filters and modulators, but nowadays this can be done in software with more repeatability and stability. However, they can also generate sound by frequency modulation which is more complex and difficult to predict. The latest form of synthesis looks at the original sound and recreates it mathematically. Imagine a piano. a string, or multiple strings under tension that get thumped by a felt covered hammer. There are also dampers to stop it ringing, operated by the pedals. The notes made up of two or three will also be slightly detuned from each other. The strings are in a metal frame that has an associated soundboard, which all add or subtract from the fundamental tone, which changes as the strings are hit from gently to very loud, and the amount of overtones changes. Each note will generate sympathetic resonances in some of the other strings, but not in others. So a clever bit of software can itemise every single one of the individual parameters and mimic them from the ground up - no samples of a piano at all, but the maths and the physics is known, so the software can produce the sound. It means that if you can itemise and measure the difference between a Steinway and a Bechstein, you can mimic it, and the good ones are approved by these firms as very accurate reproductions. You can do this with almost any instrument.

Sampling is based on multiple recordings. The very first Mellotrons had recordings of every note a single instrument, or entire orchestral section can play. They sounded good as accompaniment, but didn't convince people it was a real orchestra. Nowadays, the producer will sample individual or sections with every conceivable type of playing - so if you use a violin as example, it will be in all note lengths from staccato through to legato - and if you go from one note to another with no gap, the software will seamlessly go from one note to another sample note seamlessly and realistically. If you leave a gap, there will be a new triggering with initial attack which sounds different. Each note might have initial velocity, note velocity, pitch expression and modulation parameters, plus perhaps 6-15 different sounds. Many also have multiple mic positions. So you can choose a close mic perspective, to a very distant one. Some have advanced mic techniques available like Decca Trees with outriggers - so a string library might be 5-15 gigabytes of samples.

Now it gets complicated. Some sample libraries create new instruments by taking sounds and playing them in ways that are impossible. A harp, or piano cannot do pitch bend, but a sampler could do these things. Pianos play their note and decay, but you could have a sample that grows in volume. This becomes synthesis in a way - creating something impossible in real life. Other samplers combine multiple samples to create an entirely new sound that has never existed in real life, so is that sampling, or is it synthesis?

You're thinking in terms that existed 20 years ago, not today. Hence me keep saying it doesn't matter.

The essential features of these things has remained the same. Filters to lop off top and bottom, and filters that change as you play, like wah-wah pedals, plus modulation where you use one sound to modulate another. If you take a flute sound, which is usually based on sine waves, you can add other sounds to it and these can become very dense and thick, which is what that sample clip you picked seems to be - a rich, flute based evolving - BIG - sound. If the flute was real or synthesised is impossible to determine, but that sound is 'flutey' - clearly not brassy, or stringy.

One of the very common sounds in standalone keyboards are these pad type sounds - which again, depending on the keyboard might have their origins in a real sound, or might be totally synthetic.

I'll say it again - watch the spitfire videos and watch guy michelmore. If you took the time to learn, these things would be much easier for you too manage. One of my favourite instruments is very simple - it just combines two different sounds into new ones, you pick two sounds and the software blends them creating a totally new sound.