Truly PRO Quality AI Vocal - Fully Licensed from Audimee

[@BobDoyleMedia]

🌀✨Hey! Quick favor... ✨🌀 If you found value in this video, a simple LIKE 👍 and hitting SUBSCRIBE 🛎 helps more than you know. It lets KZbin know we’re onto something good here and helps others discover it too. ⚛ I’d love to hear what you think! 💬 Drop your thoughts, insights, or questions in the comments below. 👇 Your support means everything. Let’s keep exploring the creative uses of AI together! #BobDoyleMedia #LikeAndSubscribe #YourSupportMatters #ThankYou

[@DmitryKrasnikov]

I like how he casually treats every feature of programs (even for those witch he sees the very first time) that could be considered magical or super-complex 10-20 years ago. It's amazing how we adopt technological miracles as something usual.

[@muzy8768]

You can fix the phase issue by just moving the background vocals slightly to the left and the right plus having them slightly paned can create nice stereo effect

[@BobDoyleMedia]

I'll try that next time!

[@PhilAndersonOutside]

Used this over the past week. I sing about as well as you do, but was able to create a song I could mostly hit the notes on accurately, and cleanly. I used my voice to generate/replace with a better lead voice in Audimee, then used what fixes I made in Audimee to my own voice as a back-up singer back in Audition. Pretty effective.

[@5HTParty]

Amazing tutorial, thank you. I think we're are seeing the beginnings of a paradigm shift in how music is created. I'm excited to be learning it as it develops.

[@yinpo_music]

24:58 OK, okay okay, I will subscribe right now 🙌(hands up)

[@alwayscurious413]

In the 1970s, many feature films couldn’t afford to pay for song usage (especially in discos and nightclub scenes) so they’d hire a band to play something generic and funky as our protagonist made their way across the floor. Well now with AI they can pretty much fake the whole thing from the PC. Welcome to the world of auto generated AI Muzak - 24-7 and no licence fee payable it’s mind blowing.

[@PhilAndersonOutside]

This applies today, in all the TV shows where a song plays in the background (or sometimes somewhat in the foreground, like during an interlude). Killing the vast majority of all licensed music you'll never hear anywhere else.

[@JonathanFritzler85]

This is great. Thanks for reviewing this!

[@bobhawkey3783]

Lately I've noticed a lot of 'hash' or distortion overlaying Suno vocals. I value clarity. This video is a fantastic peek into the future of music production. Nobody does it better than Mr. Doyle.❤ Oh, and these modern slurring, mumbling singers suck. Give me Ella Fitzgerald, Harry Belafonte or basically any singer before 1970.

[@giri.goyo_yt]

"Hash" is the term I'm using from now on. I agree.

[@GodAtum]

udio vocals are much better

[@panicraptor2837]

The future of music production is "Generate Song", and "Press Play"

[@dayleywhaley2420]

Yeah the Suno vocals have always screamed Ai to me, they just sound very “digital”…I do prefer audio’s vocals

[@funkster007]

On Suno, it's more more noticeable with the latest 3.5 version. Even V2 vocals weren't as overprocessed. But then again it's hit or miss. I've done a couple rock and metal tracks where the vocals sounded pretty convincing.

[@HMaxTube11]

Bobby, Bobby, Bobby, you’re becoming my go-to channel for the latest and greatest. Thank you for keeping everyone up on bleeding edge gadgetry.🌟👏👍

[@ibdense]

Toots is in the style of late great Toots Thielemans that played great guitar but was probably better known his harmonica playing. He was the composer of the cool song called Bluesette.

[@BobDoyleMedia]

Thanks! I figured someone would jump in on that!👍

[@michaelkrogers]

gonna check that out! thanks!

[@rickarroyo]

Audimee did a very good job, better than I thought. Your voice is cool, friend! :D

[@DmitryKrasnikov]

4:50 you can put a name in a different way. You may ask gpt to describe music style of specific name and put exact description. A friend of mine got into this with 90' russian punk group "Сектор Газа" and got pretty decent match

[@scorpionleader1967]

I am getting impatient for BiaB Mac 2024. I'm sure I will be revisiting these videos. As long as you show your work (you do), It doesn't matter to me if it's a link or in the video.🥰 Just make sure the finished song has correct pitch.😜

[@samsneadd]

Hope fully one day Audimee can adapt Suno's features or vice versa so you really have one destination

[@JensBuehl]

Thanks for sharing. Love this impressive tech explained that well.

[@ZTrader_]

I subscribed already. I don't want to be pursued! Ha-ah! Great work, Mr. Doyle. With your help and AI, I am finally seeing my dream come true (I suffered vocal chord damage years ago). Thank you.

[@MCYMUSIC]

It would be great if you could give detailed instructions on how to create solo singing with choirs to your own music. Especially to different parts that have different choruses. Verses, verse chorus, etc. I did it exactly as you explain in this video. I uploaded verses and 2 verses with slightly different chords with only piano and generated vocals ;My chords are not taken over; a different progression is created. Would be great if you could explain this in detail

[@ianduck7836]

Just discovered your channel. Amazing Bob. Brilliant stuff, wonderfully explained and demonstrated. Thank you very much.

[@starholm]

I like your pace, to the point every time, no hesitation, and very educational. (Don`t play the whole song unless needed :O )

[@TheConsideredMan]

Great overview...lots to look into!

[@alwayscurious413]

Great tutorial- amazing the way you lay this all out ..

[@BobDoyleMedia]

Thanks very much!

[@raphaelbrown802]

Love your content dude, your like that dude riding the ai digital music producer wave..🙏

[@rmp5s]

Doin' some cool stuff, man! Tossed you a sub. Keep rockin.

[@mannoplanet]

BTW: "Toots" Bossa is undoubtably Toots Thielman the harmonica master.

[@IndePro-z1y]

We need an Audimee and Vocs AI head 2 head comparison. Probably the 2 best on the market right now for AI singers tbh

[@micah_noel]

Have you tried FL Studio’s stem extraction? I’ve been using it with my Suno files and it’s definitely useful but I feel like it could be better. I don’t hear much extra noise when it’s mixed with my audio but on its own it’s noisy as hell and sounds like there’s some modulation effects on it. I would love to up my game to something that I’d be comfortable releasing commercially but so far I’m just in “demo mode”. Hoping Suno gets a lot better in the near future and offers their own steam separation with high quality audio tracks. But it wouldn’t kill me to learn some tricks in the meantime to improve the quality.

[@VanessaSmith-Vain88]

Thanks Bob, you're amazing and I love watching your videos :)

[@HalalVibes_Official]

just came across this, been struggling with cleaning my suno voices i have created. hopefully this helps

[@giri.goyo_yt]

I LOVE kit-bashing! And this is the first Ive heard of Audimee. Awesome tute and demo. And I like the shoulder action when you were singing!😂 Sultry...

[@hmdhmd4887]

what is the solution with a song that have many vocals ( sometimes suno/Udio produce a song with 3 ppl singing )

[@songperformer-ot2fu]

Lyrics in Suno are getting better, but tbh, I've never used them, the latest release is good though, just want them to get rid of the autotune and be more natural, I'm a singer and enjoy taking what it creates and actually singing, they need to allow a way to remove the generated vocal stem properly, I've tried using good tools, but there is too much noise.

[@slapnpop826]

This is cool. How long of an audio clip can you upload in Suno? And will it generate a vocal performance without changing the audio uploaded too much? I basically wondering if I can use this method to create demos of songs with the AI generated vocals over my music? I guess I could isolate the vocals then drop them in over my original beat. 🤔

[@mycloudvip]

Hi thanks for this awesome tutorial. Is there a way to create or enhance clear vocals for children’s songs? I find it hard to locate resources for that sort of tracks. Thanks

[@jakemccartney1827]

bob i always love your ai stuff

[@YannBula]

Are the Suno and Udio vocals that are being extended in a similar BPM or does is somewhat oscillate? If a producer were to use this, I imagine that could be a nightmare if they do oscillate, but I''m hoping it stays constant

[@pixxelpusher]

So pretty similar to how Randy Marsh became Lorde in South Park S18E03. Amazing they predicted this process back in 2014! :)

[@JohnnyThomas-py3jv]

Interesting perspective on this topic.

[@chrissmichaels]

Audimee is the best and the most accurate voice changer.

[@neilmacmusic]

awesome job! a creative tool indeed

[@alfonsopablo]

You have great tutorial videos! Thank you!

[@penpillage]

Thanks a mill for all your efforts

[@lyricvideos-musicwithlyric4015]

You ask the question for us to comment on playing the music for us to hear. Most go overboard and play it, play it, play it. That's when it gets annoying. A little bit is ok, plus adding link if someone wants to hear the whole song.

[@BobDoyleMedia]

Yeah, I think from her on I'm going to just provide a link to the songs on Suno or Udio.

[@desventurasdeclark]

What do I do to avoid cutting the first words? I usually put an introduction with (aaaaaa), or (yehaaa), etc. It's cool to see you using Bossa Nova, for me, as I'm from Brazil, it's a pleasure. Thank you again.

[@Armones]

Wow nice.

[@davidbachy5627]

excellent work flow

[@BobDoyleMedia]

Glad you got value from it!

[@kiwigeeknz]

I wish Suno had an option to remix lyrics only. I have lots of songs I like the melody for but want to update the Lyrics.

[@morgen8250]

@20.55 lol YES. Great show tho. You are my new best friend.

[@AllanGildea]

Thank you Bob, excellent content.

[@907studiosak2]

Put in a couple oooh ooh woah's in the beginning and it will usually avoid starting the singing before the song starts, it has worked for me anyway. That's the thing I dislike the most about Suno. Cheers!

[@wancloudwonder-zw2gc]

Can you send the example here please?

[@907studiosak2]

@@wancloudwonder-zw2gc , kzbin.info/www/bejne/mJLEgaeIr5Jro6M, The next one I used a Hmmmm-mm-mm, kzbin.info/www/bejne/sKGsnXWvpdF5rsk, In this one it just wouldn't end the song no matter how many times I tried to prompt it to, so I finally had the idea to clip the last re-start and drop it in the beginning as the opener even before all the oooh oooh oh's when I did my final edit for the video (not this Suno file) in FCPX: kzbin.info/www/bejne/oZWyq3WZic-ea7c , that should be more than enough. I used to get every song starting with the singing immediately, or even missing the first words or first lines but this trick has cleared it up, cheers!

[@lyricvideos-musicwithlyric4015]

Hey Bog....I was looking at the transcript for the above video. The goal is to basically change the original singing voice into one of the singing voices on Audimee. I use Udio to create music. I'd classify myself as a beginner. Once I use Audimee to change singers, there is no way to upload that change directly into Udio, right? If possible, I want everything I create to stay on Udio and Publish it on Udio. If I understand the process for using Audimee, that can't be done....right?

[@peppepop]

A link to the song is fine with me, I sort of get the gist of it when you play a snippet.

[@Bahamamos]

how about your own voice to mold it into the song from your recording to match your voice match/adjust properly? Audimee can do that?

[@kabeerahmed7132]

Man you are doing a good job, I appreciate 👍

[@TLKStudio-xo2hq]

Hi, I'm trying out the free version of Audimee before upgrading and have a question. With the free version it seems that the only way to preview a conversion to a particular voice is to "Convert" and that uses your "Time Left", is that the same with the payed plans? Or is there a way to preview different voices without using your conversion time? Thanks

[@GodAtum]

the main problem is getting the vocal track from suno/udio, it's always gonna be poor quality. apprently you can use band in a box to convert the suno song into voice midi, which you could then feed into ACE Studio?

[@BobDoyleMedia]

You can just bring the whole song into ACE. It creates stems now.

[@GodAtum]

@@BobDoyleMedia thinking more about it, maybe there's an extra step. make a song in suno, import into ACE & do the lyrics (so the lyics are actually crystal clear), then from ACE get just it's vocals into audimee?

[@BobDoyleMedia]

@@GodAtum that would be a way to do it. But you lose some detail in ACE too, especially with AI stuff.

[@GodAtum]

@@BobDoyleMedia I guess it’s easier to manually correct pronunciation in ACE then in say melodyne?

[@BobDoyleMedia]

@@GodAtum You can definitely make tweaks on it. An important key will be sure the voice you use is trained in the language the words are in, as the will assure the best chance of proper enunciation.

[@negadan77]

Great video, like the singing😂🔥💯

[@Dave102693]

I have a question. Did you ever do music production before? Because when it comes to ai music production, it seems like you understand the music gen ai the most.

[@remoteviewingtraders3656]

Love yout channel great content 👌

[@BobDoyleMedia]

Thank you so much!

[@Dee9158]

Can you copyright the music that you can create with this app?

[@planetmuskvlog3047]

The weakest part of the project was the human in the loop, but I get your point. Ultimately nothing is happening here w/o the human

[@sagikagan6761]

Hi Bob, When you create a full song in Suno or Udio, do you have a way to generate a lead sheet with the chords and melody of the new song?

[@DmitryKrasnikov]

what is your workflow for video edition? what do you use to cut off spaces and "um... eh" from speech?

[@BobDoyleMedia]

We use Descript.

[@WillyCash69]

Now im finally impressed kudos. Wonder if this works for dialogue< seems like its cleaner than speech to speech from 11 labs.

[@BobDoyleMedia]

Yes it does!

[@Dave102693]

@@BobDoyleMediadoes it really?

[@ART-ificial]

Great content, subscribed :)

[@kevinsturges6957]

Please help me out here. I’m stuck and I can’t figure this out: suppose I created a song with a great singer in an AI program. Is there a way in Audimee that I can take an extracted track of that voice and create a model of the AI voice model….that I can then preserve and use in the future, controlling it with my own singing? Does that make any sense? In other words, can I create a custom voice model in Audimee from another person singing and capture their voice qualities?

[@wesamnadir665]

Let's hear it and link the song ❤🎉

[@Lyckantropen99]

Create the music YOU like - that is important. Otherwise you will not do a good job, I think. Your videos are great, btw. Thank you for sharing all this inspiration and knowledge!

[@BobDoyleMedia]

Thanks very much. I appreciate the support! :)

[@DjStatusss]

How do you download the video? It just spins no result?

[@davidcole2337]

Is there anything we can do with Suno finished Songs to increase Higher Bit Rate for Clearer Vocals. Suno makes great music but the Vocals are at such a lower Bit Rate you can tell it is AI.

[@BobDoyleMedia]

We just have to give it little time.

[@andreasurban6795]

The latest version of Spectralayers Pro 11 from Steinberg would have probably done a much better job of separating out the vocal track from the music and solved some of these enunciation problems. I love watching your content by the way. So much good stuff and so well presented while also entertaining!!

[@BobDoyleMedia]

My last Steinberg product was Cubase 5.0. 😄I’m hearing a lot about their newer capabilities like this.

[@andreasurban6795]

@@BobDoyleMedia Yes, in my opinion stem separation and any kinds of audio restoration tools Spectralayers offers is actually mind boggingly next level… Say what?!!! Really worth exploring. At the moment it’s gotta be the one everyone serious about stem separation should have in their toolkit. There’s some good videos already available about version 11. Highly recommend watching. I upgraded to it yesterday.

[@BobDoyleMedia]

@@andreasurban6795 I will look into right now!

[@andreasurban6795]

@@BobDoyleMedia Oh, are you?? That sounds rather exciting! There are a lot of hidden goodies to be discovered in Spectralayers. It hasn't been intuitive in the past, but I rather like the UI improvements in version 11. I do hope you find it worth your while.

[@joMojojojo]

I love when you say "hinhinhinhin"

[@mannoplanet]

where are you getting the little video in the corner? Suno isn't doing that for me.

[@BobDoyleMedia]

Suno doesn’t do that. We just add that in post so there’s something besides me to look at while the song is playing. Will generally use something like Leonardo.Ai to generate these animations however you’ll probably be seeing more animations done with Hedra in this instance as time goes on.

[@Dave102693]

Why do I hear 2 voices instead of 3?

[@BStudioT]

Thank you for your content, repeating..

[@myxy]

Nice demo. Only thing , by you singing…it kinda defeats the purpose. For me (who can’t sing at all) I’m trying to find solutions with human input. Apps like Audimee should have like a Synth v feature where the AI tries the read and convert the words or you can edit the words after.

[@toddaway]

do you like audimee better than kits?

[@MaximilianFeichtinger]

I would like the songs at the end so I can listen by just keeping it running at the end or click off and knowing no other content will come so I don’t need to find the place where you talk again.

[@stephenrodwell]

A couple of issue with this tool. If I’m trying to use a Suno song, it doesn’t do the lyrics well and takes on all the poor sound aspects. If only it had an AI fix option.

[@BobDoyleMedia]

Yes, that's going to happen when the source is sketchy. That's why this tool is aimed for the professional who is sending up a high quality track of a singer recorded in the studio.

[@soundzznetwork]

Sure, your vocals were a bit flat but you have great tone.

[@BobDoyleMedia]

Thanks. 😊

[@citadelo5ricks]

First cut at 6:00 is pure BIllie Elish.

[@peppepop]

For me, I am not interested in what lyrics that get created, more of the process of the music creation part.

[@johnmccullough4472]

Play the whole song

[@lyricvideos-musicwithlyric4015]

wow..I got lost around the point you went to another file to split things...I think. Too much information. I would say there needs to be a very simple sample (no matter how bad the voice stick with that voice. The voice isn't as important as knowing how to do it. The vide you have above seems to be for more advanced use. That's why I suggest a simple version first.

[@arpo71]

Fck. I got that song stuck in my head all day 😂

[@MaximilianFeichtinger]

Yeah you explained the extending feature this time 🤣👍

[@Melodic_Analysis]

Let's hear the whole thing

[@Gamme03]

thanks

[@reezlaw]

"Sorry for the singing" you could have fixed it i the melodyne-like tool...

[@petergedd9330]

If you can use Band In A Box you are a genius with an IQ of 500

[@cdevils]

There's a lot of mouth noises in your dialog sounds like your eating when you talk

[@quatermass8]

Still sounds like a robot singing. But in a few years I reckon they wont.

[@TheKoldfusion7]

Definitely whole song unless it goes past like 5 minutes.

[@GamaheaMusic]

I think Udio sounds better than Suno. IMHO anyway. Here's a song I made in Udio: kzbin.info/www/bejne/bqLbo4SaZsyLnZY

[@alfadelta-d1d]

Voices are a notorious weak point for Suno. Udio generates way better voices and I wonder if this workflow would improve furter the result. Anyway, I tested the Audimee service and unfortunately came to conclusion is usable only for english songs. Voices are really good and imitate well the original track to convert. But they often get wrong on some consonants. For example, italian word "CUORE" becomes "PORE" and "SVEGLIANDO" becomes "SEIANDO" (probably because in english the use of "GLI" is very rare and irregular). There is also a general trend to give a american pronounciation on vowels, so "IL VISO BRUCIA" becomes "BRUCIAIAIA.." (in italian vowels are "pure", in english are always a mix of two sounds) So, voices sound wonderfull but unfortunately they would bring a comical twist to the song... They only way to use them I can think of would be to load againg in Udio, using the inpaint to correct the wrong parts (not tested)

[@powerfunk6588]

It's obvious he's being paid to say how good this is. I tried the free version, and it's pretty lame. Most of the free voices are weak, with no real rock or blues voices. I'm sure the fully paid up version will offer a lot more choice, but without being able to try before I buy, I won't be spending my money.

[@BobDoyleMedia]

I do not say how "good" things are if I don't think they are. I just don't do the video if I don't like something. Audimee actually produces very good results with their voices. I don't know what your personal experience was or how you tested it, but I have absolutely NO issue with recommending the voices on this platform.

[@lyricvideos-musicwithlyric4015]

UDIO AN SUNO SEEM DEAD. Maybe the lawsuit has them too scared to create more updates.

[@BobDoyleMedia]

@@lyricvideos-musicwithlyric4015 I suppose it’s possible that they’ve had to freeze development.

[@londonroulette]

Get a real singer much better and more fun and if she or he is hot you can date them 👍

[@TR-ei1cy]

Started off well video went on way too long we get the point early on - didn’t need the human voice to get the point across

[@BobDoyleMedia]

Wasn't going to leave it with muddy vocals. Not everyone is necessarily going to fill in all the blanks, especially if they're new to all this.

[@dlmullins9054]

Pretty amazing, but i don't feel real emotion. Fake is always fake and i am impressed by the technology, but it just seems dead without a real writer or performer i guess. Not for me, but good luck!

[@BobDoyleMedia]

Well, the whole point is that you would be basing these conversions off of a real person with their real emotion. I was using AI as an example and my poor singing. Use a good singer who can express themselves and this will work just fine and I truly believe you’ll hear the emotion .