Clicks and buffer

[Quantum88]

I want to eliminate  the persistent, tiny yet audible clicks in  my Yeti Blue Pro voice recordings on Windows 10.

In my research online I found that there are several potential causes and solutions, most of which I have already addressed.

One potential solutionrelated to Audition is to increase the buffer size.  An AI article says :

Go to Edit > Preferences> Audio Hardware and increase the buffer size.. A larger buffer will reduce the likelihood of clicks but may increase latency, so balance this based on your needs.

Ok. Except that there is NO buffer setting in Audition, only Latency - unless this is the buffer labelled LATENCY (just to confuse everyone).   In Audacity I used to use some years ago, Buffer and Latency were two separate settings.  Go figure.

Is this the actual buffer size setting?   if so, I have it already set at the max 500 ms and yet the clicks continue.

Removing them is very time consuming, as the Diagnostic plugin is not very subtle changing the quality of the whole highlighted audio, so I need to remove 99% of them manually one by one., to preserve the audio quality.

Grateful for advice.

[SteveG_AudioMasters_]

Yes they are the same thing but confusingly labeled. If it's tiny clicks, it won't be the buffer settings - errors there are far more audible. Are you trying to record in Waveform, or Multitrack view? And can you provide a short sample of these clicks?

[Quantum88]

I record in Waveform only.

Interestingly, when recording last night I reduced gain by few degrees (from about +5 to -5) on the Yeti knob, and to my surprise, there was a reduction of clicks by about 50% . I can't reduce any further as the recording level gets too low.

here is a sample.

not sure what is the best format here so have attached both wav and mp3

thanks

[SteveG_AudioMasters_]

 This is difficult. The first one that's easy to isolate (at 0.00.880) sounds almost like a plosive pop - but it also looks like an electrical spike, as do some of the others that occur in more random places.

Now, all of the stuff I've ever seen about Yeti mics suggests that they are prone to picking up background noise, they do emphasise plosives. and that it's easy to overdo the levels. So initially I'd suggest getting a pop shield for between your mouth and the mic, making sure that you're at least nine inches away from the mic body, making sure that it's set to cardioid, and turning the gain down (which reduces the opportunities for the mic pre to distort). I wouldn't worry too much about the actual level that gets recorded; -12 to -15dB peaks is all you need, and you can normalize this afterwards to get the levels where you want them. This is a much safer way to record.

If all of this makes no difference, then report back.

[Quantum88]

ok.  I can tell you what definitely this is not:  not a background noise and not plosives.

As a professional podcaster and media producer I would never use a mic without a pop shield - althouth after years of trying and testing both I prefer a foam cap.  Pop shield is cluncky and gets in the way.

Have you ever used Yeti Pro? while they are very sensitive, from a 9 inch diustance you will hardly get a descent audio to work with.  I keep a max of 6 inches.

Everything points to micro electrical breaks in the USB port or motherboard, and the condenser's sensitivity. I can't use an XLR interface (for a reason), so I guess I have to live with it. I won't use a dynamic mic for my productions, either.

Reducing gain has minimsed the clicks, so I just need to find the right balance beteen the input (gain) and the output (recording) levels.

[SteveG_AudioMasters_]

If reducing the recording gain reduces the clicks, then there can't possibly be any electrical breaks casuing this, whatever you claim. As for the Yeti itself - there have been complaints that it has a very 'shallow' sound, and distorts the dynamics. All that points to a pretty poor impedance converter at the front end of the mic, and I suspect that is what's really happening here. Those 'clicks' aren't actually clicks at all - if you look carefully at them (certainly on the wav file) they are like little bursts of oscillation, and that points firmly at a stability issue within the impedance converter itself. It would be interesting to see if another example did the same thing under the same use conditions; I suspect that it might.

We only have one Blue mic here - a Snowball. It's suitable for Zoom meetings but I wouldn't dream of using it for anything else.

[Quantum88]

you could be right re the clicks.

When you are talking about a "shallow"sound of Yeti, that's true - but only in models other than Yeti Blue Pro.

I would never dream of using Yeti Snowball or any other model for that matter for my professional voice recordings.

I've tried few out of curiosity. 

Most people don't realise that there is a HUGE difference between Yeti PRO and all other Yeti models - because Blue (acquired by Logitech) removed the most precious aspect of Yeti Pro - the sample rate of 192 kHz.

No idea why - I suspect that it is more expensive to make it (?) and cost cutting is the current name of the game, never mind if it downgrades the quality of the recorded audio.

Yeti X has the sample rate of just 48 kHz. Yeti Snowball has the sample rate of 44.1 kHz - that's about a quarter of Yeti Pro. A quarter of the samples recorded per second. That is a horribly flat sound.  

I actually received Yeti X free of charge  - it's sitting in my drawer. Never used after the initial test. Not only it's completely flat, but to cut the cost, the mic has only the very basic sound processor inside - you need to download the software and use it , as without it this mic is unusable for professional voice recordings, it is barely suitable for children to play with it.

Yeti Pro is the ONLY Yeti - and most likely the ONLY condenser mic on the market  - with 192 kHz sample rate which produces deep, rich sound capturing every nuance of the input audio.  Irreplaceable for quality vocals.

I  just bought my 3rd Yeti Pro, and this will be the only mic I will ever use unless I find another condenser USB mic with this brilliant level of the sample rate.  Once you've tried Yeti PRO for professional podcasting, voice over or vocals - you will never go back to anything less than that.  Yes, it's got a few minor weaknesses, but the superior sound quality outweighs them easily.

[SteveG_AudioMasters_]

... and others will disagree. And if you realised exactly what difference sample rates actually make, you'd realise that if you are using any sample rate above 48k, the difference will be inaudible to humans - whatever you might claim to the contrary. If you use higher sample rates, then you will a) be creating huge files, and b) filling 3/4 of that hugeness with nothing but noise from well above the limit of human hearing.

[Quantum88]

Well, I'm not an audio engineer, but I am a professional media producer and have very sensitive hearing.

I don't claim that I can hear outside the 20 Hz - 20 kHz range, but I can hear the depth of the audio - how much information is in it - undistinguisshable to  most people. 

I can tell the significant difference - which perhaps other people can't detect - in the output audio quality of Yeti Pro and other Yeti models.   

Just like I physically feel strong elecromagnetic field emitted by devices, high voltage power lines and mobile towers which 98% of the population can't.  Can you use a mobile phone stuck to your ear? I can't, as within seconds I get a strong burning sensation in my brain on the other side of the phone, which increases  to the point of being unbearable. I can't be in a strong WI FI field either, for the same reason with increasing headache and nausea.

You've guessed it - I don't use wi fi or blutooth anywhere in my environment.

But that's just to illuistrate my point.

With the Yeti - the difference is exactly what you have described as "flatness" vs richensss of the audio.

When I first heard it I was stunned as it was unexpected, so I quickly compared the specs.  All specs are almost identifcal except the sample rate.  Given what it is and what impact it has on the quality of the audio, and with all other parameters being the same - it explains the difference 🙂

[SteveG_AudioMasters_]

As all actual engineers know, it's not what a specification tells you that's important - it's what it doesn't tell you. And you can cannot really determine anything about the behaviour of a device from its specification alone - there are many, many examples of this. It is even possible, in this instance, that reducing the sample rate from 192k to 44.1 might actually help your situation; stranger things have been known. And that's the point; you can't determine what will make a difference just by looking at a piece of paper, especially if you haven't been trained to interpret what it says correctly. And in this case, you need to be aware of Shannon and Weaver's work, and Nyquist limits before you even begin.

[SteveG_AudioMasters_]

Let's just look at what it is you can actually hear: (This occurs in a lot of places in your file, on both channels - there are many more)

The bits I've ringed in red show pretty good examples of instability somewhere in the impedance converter and preamp stage of your mic. The problem has to be there; there is nowhere else analogue in the mic before the USB conversion, and if USB converters did this, we'd all have been up in arms about it years ago. This sort of instability can be 'provoked' by having the gain turned up too high, but there are other possibilities too - mainly involving poor design. The relatively smooth part of the waveform is actually you speaking, but the waveform superimposed on it gives the whole thing a 'grating' sound.

[Quantum88]

NOTE:  This is my reply to your previous comment.  The one above was posted while I was still typing and before I posted it. Yes, the impedance converter's instability is a possible cause.

I disagree.

I go by what I can hear first and then look for the explanation in facts on paper. I don't need to be an engineer, have a PhD in electronics or be trained in interpeting complex scientific data to know and understand what I can hear.  I also know what I know even if it can't be scientifically proven, and I stand by my conclusion.

 I do hope that it will help others looking for solutions here. and encourage them to test these mics themselves to hear the difference.

Anyway, the sample rate was just an add-on point, not the key topic in this thread about clicks and buffer. It has nothing to do with the clicks; it was just my comment on your point about the "flatness of Yeti mics" to point out that we can't generalise as Yeti Pro is VERY different and of superior quality to all other Yeti models.

I have found the (potential) cause of those clicks, and most importantly the way to minimise them, so we can end this conversation. thanks.