@Lindon I think @Dan-Korneff uses iLok on his plugins

@David-Healey Probably, are there any examples?

@David-Healey said in NI Insolvency:

Would that need to be part of HISE when other tools are already available to do those things?

Yes, like scriptnode, or sampler or synth stuff in Hise, like other tools are already available to do those things.

@David-Healey said in NI Insolvency:

@resonant I'm not afraid of it (I made a whole HISE AI workshop). I just see a lot of vague comments without any specifics which is why I ask.

If I can write a prompt and a plugin pops out then all our plugins are worthless because the end user can write the same prompt I can.

What I'm referring here isn't creating plugins with AI.

As said, it's about using AI to analyze and then using DSP to process or generate DSP in a shorter way. For example, removing reverbs, or properly removing snare drum from a mix, or creating very realistic physical models…etc. That’s what I call AI Workbench.

@David-Healey said in NI Insolvency:

When we get to that stage all our software becomes worthless.

Whether you use it or not, definitely someone will bring it to that stage anyway, but you don't need to be afraid of it, because software won't become worthless.

It will just work in a different way. Do you want to be ready for this stage or not? Don't we all want tools that make life easier? Besides, AI makes significant sales increases.

@hisefilo said in NI Insolvency:

This is the end of an era. AI will be the standard. NI didn’t realize that. Hope HISE does. ( by adding deep integration with AI frameworks. Dataset processing etc)

No doubt with that. Although we have friends with opposing views, I agree with you on the same point.

I hope HISE will be an AI workbench.

@Christoph-Hart Have you had a chance to look at this? The last fix doesn't seem to have worked. Maybe it's something simple?

@Lindon That was funny I hope this stays true for many decades to come—I'm definitely not an AI fan.

I think what's being overlooked here is focusing only on current capabilities.

Let's not forget that this is just the beginning. Think ten years from now. Consider the future of machine learning, the DSP code that will be produced with AI, and other features.

Yes it is not today, but it will be in the future.

@David-Healey it should be one day, but not today

I just saw this. An AI-powered audio plugin creator

https://freqlab.app/

https://www.linkedin.com/posts/jamesontucker_i-wasnt-planning-to-share-this-publicly-ugcPost-7416904104506966016-Rzq5?utm_medium=ios_app&rcm=ACoAAAyCtrwBFQ8mktWhposHpUR7JcYSDZMrStA&utm_source=social_share_send&utm_campaign=copy_link

@Christoph-Hart said in Neural node bug on HardocdedFX as embedded:

@resonant fixed now. The problem was that it tried to load the network in the hardcoded FX before the first script compilation so it couldn't find it and there was an uncatched exception.

Unfortunately, the same problem persists. Hise still freezes while loading the project.
I've just compiled the latest commit with the latest Juce 6 submodule.

@JulesV Here you go:

HiseSnippet 1231.3ocuW1sSibCE.1Cvrhj1cKqZUUubDWEVnnLjDftAzll.zF0BjtIPqzVJxwiCwhI1Qy3vlzpUp8MnOL8IoOC8JTunuAsGOdlLSfrvRR1NWM97iOe93i+qlmfP88EdHizMFzkhL9Py5C3x1UZiYbT08PFOw7Pruj5YoEUdPWruO0AYXL+WoDXjZATv2+7hxXWLmPiEgPmJXD52x5vjwRqU5aXttGfcnMXcRXc9RUIBdEgqnGvy7lYQcwjKwWPOBqLaNSzWi8aiLdlIwIOIOYyl4sw3VaY2bSZyss2xltMofyWPHXRglD6b4vHiGsuCSJ7pKwRpOxXgxBmA0aKdMWGfSY9rltTUCaTcHxZwGHbcTCQkTTk1LWmZQIJeDzK0hSayqSaeh4gLG1P4woukBTXE6QxDnwbih27ifmcR7xl.uwfjQBjVPizSMqS7XckwZT77AlU4vrYKLLOkDEssn49zELqH.K3x06fujdfGzXnGYxWH6ZV4ylckhoSmFlr7kVWg8rN5jCOu7IMZb7Q0s10pPwDppAyek6Ikf.P0qNKRmRdMLm5FINTd4y+dlirMHbyrESJiNTgc1rpv2R3YkgAsyVzhYsSRH.Aqt5Jo+kzVvWbjdE6Lv7ng2ETYEQmtBNzHyxAVrr0pPWtp8Jv3S45P+CGA2UGnM4l8vnNutuxEtzS3VA651DJuyH3A1Hk7PMfmuIs5iwcYbpUqdbhjI3V2xxLjnvuFjqc6QUC4TtBB10h4zGPMQ3Wmwcn8OtUrSPfRkJcpalGiyWq6R4WHaGlMUCGcF8VY008gZ1HnBh7tVrvLv3yBmpvUE0cUjlvz6vmLK2e40Tnti1IqW.0HP9Npj4Y.9OWKJyHkLedjIqLp0qTL06TXesxAcn2cXnGIBOeXDziDXB7Mou87UxNeiyVyR5AyYESG+kXt9HgjdLOSvTJzYV2TUqViUWXnbodiUsZGWu6xwL7dcZR8RVPoLD1UYzspLe6aUkbmThdsRBCE7pbl73tT9aa+UT3BL3uSptGVhU6uEJCrqK0SxTHXrG8J3zE8tcoL2i5eoTzENIa3pRjwhx.sOIbuPcxGwbTNnaXi5CMdzNgGXMX34Q+52UJXhOo11T1EsSbT1dkjz9R8.wGeEsJulGEpYT6nhdPXrgFi9SIFaLkXjSiwebmX3buXjaJwHuFiOyX5xF4mRLJnwnwThQg6DizZLdbzw0pMUCnXQyf+0EngcZiRwPf9ogPnETRT5FPTpaIlj1I5NUodzRKccVkhlBOGpWc1OOxs0zReI1g0yOo7qhVb+vHeiYI4eze82+9Li7T2C44lkj+mhOdvCm7GRhN+LsD45co+uUhTXVR90+1R+3zQ9suEMb+cgSOWrbzK0qd5SnB33yQtIs51xbelbPxmFMyto+6JhO0rFSRZOdFmaLLBGx99fwv2G8Xy8a0hRjw.tf4A+v6mGCgdonmjwu3PrziAkWlG0qScnDgPgnyUWYEjYLmZIltcVUaUFnNk6Dz3eguPk1p1FgJsiTh5fIdhyI5KModA1hAR.l3AuVME7rYnskMJ3hTIyycfGDdNgLZWcKG2XRcL2j5X9I0wBSpiaNoNt0j5312uip2q+k8jhN5kMHzg01O3VnFF6ywPEXP0J5+.goKbaA

@dannytaurus Your concern about not upsetting subscribers is justified. But you can overcome this with a good email marketing plan. People who don't want emails will unsubscribe anyway.

But I think keeping an up-to-date and active list is much more important, so you need to manage your email campaigns well.

• You can send monthly discount coupons (every month) to those who make purchases. Additionally, a one-time fresh thank you coupon code for users who make a purchase would also do the trick.

• Flash sale emails (monthly or weekly if you have too many products).

• If you create blog posts or videos (useful content for users), emails related to these will get good engagement.

• You can send emails encouraging reviews (even offering discount coupons in exchange for reviews).

• You can also send special launch campaign emails for new products, user reviews, etc.

If the email content is somehow useful for users, I think you shouldn't worry about those who unsubscribe; as they are gradually eliminated, you will build a core, active audience.

@Lindon said in Taking my Mailing list seriously, so a first question....:

@David-Healey yep just found that myself... loks like I will need to upgrade TheNewsltter to Professional ...

Yes, you need to upgrade to pro. After upgrading, you'll need to install the Amazon SES plugin too and obtain data like the Access Key ID, Secret Access Key...etc. from Amazon SES.

The installation is a bit cumbersome, but once it's installed, it's definitely worth it.

@Lindon I haven't used Sendy yet, but I am using Amazon SES. You should use Amazon SES too.

AFAIK, it is the cheapest one. Sending emails to over 100,000 contacts costs over $800 with competitors like MailChimp. But with Amazon SES, it might not even cost $20 :D

I am combining Amazon SES with the Newsletter plugin. But as @David-Healey says, Fluent CRM is better.

@Christoph-Hart will the neural node bugs be fixed with the Hise 5 release? Please, at least this one, you can try the above model and see the issue.

There's currently a bug in Neural Node.

The model I embedded attached below was working on the HardcodedFX in previous commits. Now the Hise screen freezes when the project starts.

Since the model is long, I wasn't able to paste the snippet.

NeuralAmp.js

@ustk

I think the calculation can be done using the float getSamples function and then passed to the mod dragger. I managed to do it up to the code below, but the rest is beyond me for now.

HiseSnippet 2980.3oc6ZzzaabbcojFaSp3D+U93PNLPs0ktUgTTVwwA1xj5yXUKJQnUVV4jvvcGJtPK2YytyRKl.Czi4VN6Con.4WQOk7GH.EHGxoBTfdo8l+Gz9lY1k6rTTVzJ1pwAYkfzNuYl22u2LuYV5gbpmM0duPOGeeJ2vH23MBXVzvPVfQtBa2ymZj6MPl873sWpMwwyXskMxcEzRrf.pKg6v7pS4z.iE64SBCo1F4xM9mHFWt7SXHedV0EItDOKZJHCicXNVz0c53vSg1n1CbbcWkXS21oi1nmq1ZVLukXtrHfmFGMigOw5.x9zMHhgMFx39jv1F49CHK64rly5VMmqBgz5ipz7Vzl2txGUgdaqOz9isrHVeXSqJ27lDibmaEaGNKvjS3zPibSrHytmYa1i8TDXGmPmltTQiJFl.kUfWk4ZKDQATikZ63Z2WYEBptyoo5FWo5tFptisSe3opvKI6.mNCcEXtwxxdimg8pnydynwdCgkxowRSnXoKiLsBb74o8H3mIQq4A1wVDvNoyJpwZj66.SNL.OdoNjCnqF.M5Ogh2ZlYlFC+4F2oUjmkvo.y71fwoa5U7FE9hB4K7jB3A6pUqg1mfLALWWZvP6V3ZD77lXQunNMoASi6Rbin8GHH9Y0oniWmpaxsTRs1.Ydq44v2zm5cbNBFwpJ3sGt1xDNQXHhgAiymFvcDrPtkocgv.kYIOZYZ3AblODHbDaF3svrif3srtPh.s3N.cPF6lv33E5v6oGH9RyuZTYwKiZ3vsZObdbrgvifl5UAOFGMdQzJsZQs3oL3DnU28rKzqPbnGPSAwuHxbCQhz57fc2cnQceIJ0C2Of5SBnayZ3R5ULjzw2ktEvjSia5xrNvz4yoGMrvWwCKJFQQq1DOOpa3oI54birZoxy2prEKh63secBOv4PHLbinNlPhcK5RwbG.K2XhfHU6YDsENFlvJUxF+W3ItyJh14h6rRRmZweaP4OlEbfzdD+tQtKHT9EPIpdiEbcYOdIVGemXmTkHQCBkMNOZlRvOpg0f41yuMyywRXmTSJg0WnCKBnZL+m5zuXh8IQTf0q1l33JhALiBgLI1a5YB3QtHovGbClM714VkXA57dMH71hXIQtHvwkFTxpuCrlXj5.IT8pEG05VfyjEQUrAv8cDpIwRn4FWzuPOMdLwEX+pHpWWpKjxZuVLg3CqzmgodejcOORGGqvRGcj5dzJ1YX36D3qIzxXpqT2lDrOkGFmZ0iZorbRU2Z1hEzB8nGtmmPXZPB.xCqVI537HGO+HtjLPHzgIInW1IzGhrVLBRQnBfy.wPjuxldXhS1JPbhMra.4zSb75SHohreKvNT2waGQHkdpt5jCyBq1WUyjS8UtJRX+0m9z+VUyCnOVo0igh+9u96qJrOW.s.mCaFxXa5g7sYRrAZitf9RP0ygj6jpArWIoxzLp4mDvh7kMzHcsCZ052TaYZKRjKe.1btZFu5ki7nsntTRH8mnf3+Wd5mZ9RTPf0LGPPTvFTPjPEBxaghS1Y5ruGw8XDmIQODbkZAgx1mrLodFtLIxvkFsp6xmIJ8RPpi.ZozdGVpB8Y+hDSJWnSzpmHDQZLWxEvNDqEys8SBE7hD.+lH+Nj8h7BsHtR8TlXXEpGHFFsMAVDRjmvz0wlFz.BMDsVHx1gspipK3+p.TiAi3yF9lSya47odKHwpAZdJHTEnsPtSRrrPDm0AVlTtGmT2G0D0ccTSMikMi8b.MPFS56HWM.VwtTlwbzTtCfjSZUfSJa6wmh6MNZD0y91QKhxv33inRL0uxijVSgTopKiI7DRa7xPHKfpCrjiuauWVx4ypNb47YUOCDGDZAa6yfbe59hFO2bRuAZsvcD8BQAID.rrKxXGzgH2e3O4R0tZRMFxrMqyDSVlW3ZJHXAHrXy6zfJYpC5TddI9i74kX1yyptLuRJQjElsc.jgD1MtNoDr4JdhDo15ieKFiKpaW+7ZLkEg.0DDvAzqijsnf8NjlAlHArJmzWiJuCyMpCMrrPmWdYPS2jcX4MV3Qkq+Py0Vpr4B0ar9J30WawsVXqOsbiEV5AlkMi775gWmbXYStTitBnA83ND2PrBe3YKCFuPdPjptlG3vwe.dlJyT1bcyUvyhGdu3Ue35qiqu1tkdLoqQGg0.k6o+ieuQGhb+de0aujgKnVjRZRetR8jcR+4F+4cjRUNkGozD+r6HkdM3zIxT5c9DdzzQ3rtRb8GwmiYbpkDnYhKGuNyikTqmtqMT4596Syl0aXBT7dy6C4Z0R1kaBnu7GpsNjIjDLXz4KftnxK7wPLT606EWJAVjLA+56oJM9unOUoyjiOQSDeWkHNI5QjtT4s.HEv2V1tEKnC9Sndz.g8oxwcG.+4dW8u+xeMsMs3.K.Kf4E5yBgH2TL+clzNNay7ng5PGbFypumm9yX1ra7fG4kA0wfxLJgpXUPUnMtKTKA3r5.aP7xfLncFLUWXRSZ9OqpVE1jBRt8lgVfhgLvlsZD4FRejiMucE8IlBdVcv2mDXK1JPl.iWIKa8y7aBoef6BA9z822QpVUUboAIqC8fJpS+If1+zIyc03ScEFJWxQuY7ADuXDmCo0bjkAq3hXPcOxlfGAbbQzVzPJ+EFEppnkn3sjYZDUAnfM.RduZiBRlDIvfjWNcbwkjHv7.migMfpaFA0AfEV.+Azdg+TzHfv3So1Ovi0bPDb8ipMtTLBfpsaxVjcnDEWAYR+rHww91G7nKPZrxEQpbaaQ7fBpNURSdjY6nVsbomNOrIQpjUa6PedhvDJDbkLzWbJMRjbMTRZXsNF.aWqls7.W+cnYmqz.OM5+1G+.3O2uqwnS4cf8+YAaZ33o7y91LTt0ebwEDDZ4RktYTiAaOxT9Jn0od6yaOxRbqCDzQ9rKquDuyeBnaI6FFili2kQw6.vD1PprB9S1uanH5cQaFw8i3ujP2kQ0arhrJqXDdDuwQLx5sDHBpN6XPyL0FEm5ygtOrL2PBH9kbIPydFWBjwu8WKA5WKA5Ew0dLM5Ooh9mG7hcI8LT2Il78r6dSBRXVWm1Rqbi2uV+N1xY+1bcG0UgE2aBdCZyn1Mu4OTMAt9DjcrN6wM.BtZ.8yzKv+9NGApwx0R2pesB+3OTcaZGeV18lCdUi+5zktezDNuYbBG4gA1LNknBDVAKqQZKFqS16L7+TcYRG+zCnDzy+qpOhxWGlsaLr+423+8UWNnmNLie7a92UkE.oiKvFP+bpVdF4sR9ZqV94uzC3f8y6TXvVGBslEKyig++ShryBC1YAM5PrBX6EegfhTjWPBAjaO44smGUWzFW4n6koCT68dVVYQ0Ql3rm1IdyS6Dm6zNwO7zNwacZm3GcZm3sO4IJNZj3KiTDaXX.aqU8Emjq+czjabCZ7GihZKrBW69P7nGFey24FqkiHrP9OEm7No24e4j6nuT6jO5QiwlYLDGVYBhLo3653wwary8vpKPAmLdw2Hl4Fqr6dar4xqTLA5MtSg7EJj2lEALId28wyiqTZFAP3mlLlKF7nscooA9EUi853t2nPd.m4U+1ElID7ztTfn31hcwkArMM9C.rMs.kRBkOe.kGE3gAViB.dhj3YY8F2C2k4XiCo792aXLM6SRmVEafmed7L2.WtLdMw04CPkbe26HXGAhyKQih8uuSHcktftpX+2tNldiu3IwiBH1JGBDxi3JLaEsD2+DVGz0w1SiE7mi3qKJclweteEan9uoO0JD6GlNf.wwnTT1NUPE0xHLsIGmlfBhbvwxd7mCXwA5E3A3sAPkfmVJ67VUn1JF5S7taKWFgOMLf9yMegmbmBFFGwG68FlO1gcbS8xlLG5tUAH3tpO4t4mBrqSgodVLaH047S8vsW8Ct8TUuWgB2M8JeuWALNsIN4lrmepYDSN4RriwUx8WG2a5UWq5GPk1yZKO+TxOkiovC6NrA9I4FrmBKtB64mBvX7MXCumEY5Lk9EYq.U9dEtaYcYxv3+A+Zd6PB

How can I make a node using SNEX or a C++ node with modulation target (value range between -1 and +1), just using the useful Correlation Meter code below?

(By the way, Thanks to @dannytaurus for providing this https://forum.hise.audio/topic/14007/phase-correlation-meter/10
)?

function processBlock(channels)
{
	var left = channels[0];
    var right = channels[1];
    var numSamples = left.length;    
    
    correlationSum = 0.0;
    sampleCount = 0;
    
    // Process each sample pair
    for (var i = 0; i < numSamples; i++)
    {
        var leftSample = left[i];
        var rightSample = right[i];
        
        var normGain = Math.max(Math.abs(leftSample), Math.abs(rightSample));
        
        if (normGain > 0.0)
        {
            var thisCorrelation = 1.0 - (Math.abs(leftSample - rightSample) / normGain);
            correlationSum += thisCorrelation;
            sampleCount++;
        }
    }
    
    // Average correlation for this block
    if (sampleCount > 0)
    {
        var avgCorrelation = correlationSum / sampleCount;
        
        // Apply smoothing (same as backend: 0.8 * previous + 0.2 * current)
        smoothedCorrelation = 0.6 * prevSmoothedCorrelation + 0.4 * avgCorrelation;
        prevSmoothedCorrelation = smoothedCorrelation;

        corrCable.setValue(smoothedCorrelation);  
    }
    
    
        
}