Can I make the words any clearer in Tal Vocoder?

[DallasSteve]

I've got the Tal Vocoder working in a song I recorded, and it makes my vocals much more pleasing to the ear, but the words are not very clear.  Is trying to mix a copy of the dry vocal the only solution?  That will bring back what I wanted to avoid, but it may be my only choice.  I've tried playing with the sliders and buttons in Tal Vocoder and I haven't found a setting that helps much.  It seems like the best setting is to move all the Carrier sliders up all the way as shown below, but it still isn't very clear.

 

Edited February 28, 2022 by DallasSteve

[Byron Dickens]

Um... That would kind of defeat the purpose of a vocoder....

Edited February 28, 2022 by bdickens

[DallasSteve]

12 minutes ago, bdickens said:

Um... That would kind of defeat the purpose of a vocoder....

Is part of the purpose of the vocoder to make the words difficult to understand, because if it is, they succeeded?  In that case, has someone invented something that does what I want?  We could name it the vocoworder.  I've tried using Melodyne auto tune but my old voice is so bad it I can't smooth it out with auto tune at max.

[Byron Dickens]

I  believe one can think of it as using the voice as an oscillator for a synthesizer.

Edited February 28, 2022 by bdickens

[Lynn Wilson]

Try sending your vocal to two different busses; one with the vocorder and one without.  Blend the two together, adding just enough vocorder to get the effect, but not so much as to muddy the sound to achieve clarity.

[DallasSteve]

1 hour ago, Lynn Wilson said:

Try sending your vocal to two different busses; one with the vocorder and one without.  Blend the two together, adding just enough vocorder to get the effect, but not so much as to muddy the sound to achieve clarity.

That's what I suspected.  I wish there was a way to use the same vocal track twice (once for the vocoder and once for the dry vocal) so that I don't have to make a duplicate audio track, but I haven't found it.  I also found a way to improve the vocoder a little (perhaps).  I was using the internal synthesizer in the vocoder which is a traditional vocoder sound. Now I am routing the signal to a synth in Cakewalk and I am testing the different voices in the synth.  Some voices produce a more intelligent speech than others, although it's still not as clear as a human voice.  I may do as you suggest and feed in a little of my dry auto-tuned voice along with the vocoder synth and I may be happier with that than just a dry vocal (if it's enough to be worth the extra work).

[abacab]

50 minutes ago, DallasSteve said:

I wish there was a way to use the same vocal track twice (once for the vocoder and once for the dry vocal) so that I don't have to make a duplicate audio track, but I haven't found it.

There is.

Read Lynn's post again. You can do that with send busses.

On your audio track "Insert Send" > "New Stereo Bus". Do that twice. Now insert FX on one bus and mix to taste.

[Byron Dickens]

Based on your "old voice is so bad" comment, I think your real solution here is voice lessons.

Edited February 28, 2022 by bdickens

[DallasSteve]

3 hours ago, bdickens said:

Based on your "old voice is so bad" comment, I think your real solution here is voice lessons.

You haven't heard my voice.  Or maybe you have.  I posted one song here.

 

[DallasSteve]

3 hours ago, abacab said:

There is.

Read Lynn's post again. You can do that with send busses.

On your audio track "Insert Send" > "New Stereo Bus". Do that twice. Now insert FX on one bus and mix to taste.

Thanks.  I didn't understand that's what he was saying.  I'll give it a try.

[Byron Dickens]

18 hours ago, DallasSteve said:

You haven't heard my voice.  Or maybe you have.  I posted one song here.

 

Just for that, I went and found it and listened to the whole thing.

1)You're singing flat.

The reason people sing flat is typically because they are pushing too hard and/ or there is a lot of tension in their vocal tract.

I can hear the tension around your larynx. I can even feel it. If I imagine myself making the same sounds, I can actually feel my vocal tract tense and strained.

You need to relax. The more you try to force things, the worse it gets. Counterintuitively, the higher and harder you sing, the more relaxed you need to be.

2)Your performance lacks passion. You don't convince me. It lacks passion because you are trying so hard not to mess up that it gets in the way of conveying  emotion.

And that brings us back to 1: relax.

No amount of magic plugins will ever fix that tension or that fear.

I know that we have gone around and around with this before but I'm telling you: if you spent the money and effort you have on plugins on vocal training instead, you would be much better off. But if you have already made your mind up, nothing I say will convince you.

I didn't hear a terrible voice. I heard an untrained voice.

 

[DallasSteve]

bdickens

Thanks for taking the time to do that analysis.  One thing you may not know is that recording you heard was after throwing max Melodyne autotune at the vocals, and it still sounded shaky.  I'll try to implement what you're saying, but I still don't think you appreciate the damage time has done to my voice.  I have access to something you don't.  I have some recordings of me singing some songs 30 years ago.  The difference is huge.  My voice cracks a lot now in addition to having much less range. 

I've listened to Paul McCartney's voice for 50 years and the change is also enormous.  He used to have an Elvis level voice, in my opinion, but now at almost 80 it's falling apart.  Some of that may also be cigarettes, which I never smoked.  But the point is, I don't believe all the vocal classes in the world can get me  back to where I was 30 years ago.  I do have a couple of other possible solutions.

1 - I have a 25 year old son who may be able to re-record my vocals.  Who knows.  That handsome young man might become a star.

2 - I think I've found a better vocoder solution with Ovox.  I'm playing with that this week and I'll post an update here when I've reached a conclusion, but it's looking promising.  I can sidechain my guide melody into Ovox and my vocals sound spot on and not shaky, but the words are still a little fuzzy.  I'm working on that part.

Edited March 2, 2022 by DallasSteve

[John Vere]

He's my favorite Plug in for smoothing my vocals on a day when I'm struggling. 

There is a lot to be said for a bit of vocal training. I sang for years without it and sucked. Good thing in those day's so did PA systems so nobody really noticed. But then there was the eye opening day I wrote an original song and tried to record it! This is in the days of tape so no digital turd polishers to be found. I went to a local piano teacher friend of mine and in a very short period of time she had me on track. I would assume there are lot's of tutorials available on You Tube that teach the basics for free.  

A note about Melodyne- It can most certainly correct minor and even major pitch issues and if you get good at it you can do all sorts of wonderful things like split notes, fade in's etc. But you have to have a light touch with those tools or your track will start to sound like it's been through a meat grinder.  It's taken me 2 years and 100 hours to start to understand how to use it properly ( I'm hoping) 

But absolutely the best approach I find is if I'm fighting a phrase or line it's best to grab the Glenfiddich VST and re do it. 

[DallasSteve]

I have worked some more with OVox and that is a good solution for me.  My final solution is to do a mix of the original vocals and the Ovox synth.  But I also used the MIDI notes in my guide track to pitch correct my vocals and it works really well.  I realize that the final vocal has a robotic sound to it which can be lessened by reducing the synth in the mix, but I am very pleased with the result.  I can improve the result also by adjusting the placement of the notes in the MIDI track as sometimes they don't align exactly with the vocal track or are too short and cut off a sung note.  I will polish that later, but I want to upload this new demo first.

I have created a new version of Don't Kill The Unicorn on Bandlab as a demo of this process.  To me my vocals are not what are important.  To me what is important is the composition and the arrangement.  To me the vocals are just another instrument.  Now I can post this on the internet and invite friends or the public to listen to it and not feel embarrassed by the vocals.  I know that bdickens and some others may not like this result, but I am elated.  I feel like I have solved the final hurdle that was tarnishing my finished songs and I am very pleased with the way my new album is sounding.

Edited March 2, 2022 by DallasSteve

[Starship Krupa]

One thing I figured out, it seems obvious now, but my voice is an instrument, and as such, improves with practice. We didn't sit down with any other instrument and expect to be able to use it correctly without practice.

I learned this by doing take after take on a song and noticing that as I went along, my takes started to get closer to being on pitch.

Well, duh. Any instrument needs practice to achieve proficiency and the voice is no different.

If what you want is the obvious pitch-corrected robot voice, Meldaproduction's free MAutopitch is a handy tool. I put it on in front of the vocoder and they're a good match. It alone might get you where you want to go.

I made a tutorial about using TAL Vocoder, it's over in the Tutorials section.

[Byron Dickens]

Honest to God, the vocals sounded 100 times better before.

[Byron Dickens]

[John Vere]

What’s important here is you’re happy with your creation. If your goal is to make yourself happy then you've achieved that goal. That's awesome. 

We all have a standard of quality for what we create. 

That standard is based on our expectations of what we want to compare ourselves to and our ability to achieve that goal. It also gets higher and higher as our abilities to create improve. 

Recording music can have a very high standard if we wish to sound like a hit record. For many of us this has been our lifelong goal. I want my songs to play side by side with commercially made material.  I most certainly realize most of those recordings where produced by some of the worlds most talented musicians and studio engineers. How can I even begin to think I can achieve this? Is that too high a standard for a small town guy in a Home Studio with a Cheap guitar and a computer?  I'm a Capricorn,  my standards are going to be a lot higher than others and I'm up for that challenge.  

Bottom line is, each individual will set their own standards and be happy if they have the ability to achieve that goal. 

Your song is so far below my standards that I dare not comment, that is not fair, I have a lot of trouble keeping my mouth shut on the song forum. I need to remind myself of what I have stated above. I truly love this hobby and want others to join in and enjoy it too. I would rather offer encouragement-- that you too can raise your standards if you keep at it. Take this song and file it away,  keep creating more songs.  Crank them out and keep studying the art of recording ( and singing) .  A few year's from now your talent and abilities will greatly improve and you can come back to this song and re do it to your new higher standards. Don't obsess too much over your first creations you'll get discouraged and quit. Just keep at it.  

 

Edited March 3, 2022 by John Vere