ohhh LEVITATE took me 5 mins to realise what she was singing. "I think 2024 is going to include a lot of AI. Some of it is going to be trash" lol quote of the month
The #1 issue that this target market faces is that this only works (assuming that here), if one already can/will sing the song/s. Then, of course, the AI "singers" sound emotionless and like they are MeloTuned. Admittedly not as terribly poor as this sort of stuff: ruclips.net/video/DBOw_LWJgMA/видео.htmlsi=n_FjDG_zT3XHLmqR BUT still not remotely the equal of a passionate singer - even if that singer is imperfect. It will only change a certain part of the music game, i.e. Gamers as real Artists and Fans will not want this as it is an abomination. :-)
Suno ai singers are absolutely not emotionless. In fact, nobody who listened to my Suno ai song thought it was an AI singer. Everyone thinks it's a real singer. But Suno ai is also on another level. These vocals also sound pretty good. When you export these files, they are never masters. They are basic tracks with the dB kept low enough with plenty of editing room. After running denoise and mastering the tracks, I get them to sound about 500x better than the basic track. Also, suno let's you add lyrics and get a complete song with vocals and many times, their vocal arrangements are absolutely brilliant. Yes, you do have to waste a lot of generation on stuff that sounds bad. You have to keep making copies till you get a good one.
Because of technology like suno ai, it's only a matter of time before an ai singer has a number 1 on iTunes. I'm sure that will happen within ten years.
Hey Jeff, good video. I think you'll get much better Audimee results if you would fine tune your own vocal performance (With Melodyne or Repitch etc.) before processing it through Audimee. This groundbreaking stuff for songwriters/ producers who can do a good/ great vocal performance but simply don't have a world class voice. Audimee will do wonders if you can get a really good vocal track of yourself and then take it to the next level. For female vocals, you can also pitch shift the tracks in your project (minus the drums) a few semi-tones (like 4-5) and then do the same with the vocals. (Or built your tracks in a higher key, then pitch shift it down temporarily to add your own vocal performance. There are many ways if you plan it carefully.)
I wrote a song that I felt like would be much better with a female vocal. I used Audimee to convert my vocal and I'm very happy with the result. Keep in mind, I started with a very clean vocal. The only processing I did with the original was a little compression and tuning. I did have to transpose the song because a female voice in the key that I sang it in just didn't sound good. I used the key I wrote the song in when recording my vocal and then changed the pitch to +3 in Audimee. The converted vocal had the same feeling and same emotion that my original did. It works great.
just to clarify there is nothing free, independent artists and without conditions audimee is free only for feedback, ideas and training, it's just to try the free plan does not allow you to use the vocals to distribute your songs, it's good to read the terms before using or distributing your music, however I can say that it is on a good track and works well with small and not so serious errors!
Great vid Jeff and thanks so much for sharing. As a musician who only needs some backing vocals on my songs, I think the free version suit me just fine. I'd just be uploading the chorus lines from my songs and picking a female singer. I'd never need the cover version section.
What happens when you use up the entire limit - 15 minutes, in the free version? Will I have another 15 minutes next month? The reason for my question is that the word 'monthly' does not appear in the free version.
So inspiring. I have been doing the same thing - I’m a songwriter with an ok voice and I’m enjoying using it. I have the pro plan and am trying different techniques to actually get a pro, usable vocal where the artifacts are not too bad. It depends on the model you use and how you record and articulate your words. I’d love to share tips snd tricks as I’m on a mission to actually get something professional quality - on rock stuff with a denser backdrop it is passable but at least with my voice I haven’t been able to use it on quite passages because of noise and artifacts. I input clean vocals from my studio. Anyway, thanks for the video! :)
This is the first video that I find that shows results on custom voices. Thank you very much. It would be nice to see a good workflow in order to get the best results when training vocals
All AI tools are a little scary, but I find this useful. Remember, it's not generating the vocals. You still have to create a vocal track - but then you can pick a pro voice clone to take it to a higher level.
Today it’s a tool. Tomorrow it’s looking for Sarah Conner ;).
Great video !
LoL Love it!
ohhh LEVITATE took me 5 mins to realise what she was singing.
"I think 2024 is going to include a lot of AI. Some of it is going to be trash" lol quote of the month
The #1 issue that this target market faces is that this only works (assuming that here), if one already can/will sing the song/s.
Then, of course, the AI "singers" sound emotionless and like they are MeloTuned. Admittedly not as terribly poor as this sort of stuff: ruclips.net/video/DBOw_LWJgMA/видео.htmlsi=n_FjDG_zT3XHLmqR BUT still not remotely the equal of a passionate singer - even if that singer is imperfect. It will only change a certain part of the music game, i.e. Gamers as real Artists and Fans will not want this as it is an abomination.
:-)
Voice cloning is here to stay. And I'm sure human singers will not be totally replaced either.
Suno ai singers are absolutely not emotionless. In fact, nobody who listened to my Suno ai song thought it was an AI singer. Everyone thinks it's a real singer. But Suno ai is also on another level. These vocals also sound pretty good. When you export these files, they are never masters. They are basic tracks with the dB kept low enough with plenty of editing room. After running denoise and mastering the tracks, I get them to sound about 500x better than the basic track. Also, suno let's you add lyrics and get a complete song with vocals and many times, their vocal arrangements are absolutely brilliant. Yes, you do have to waste a lot of generation on stuff that sounds bad. You have to keep making copies till you get a good one.
Because of technology like suno ai, it's only a matter of time before an ai singer has a number 1 on iTunes. I'm sure that will happen within ten years.
Loved your summary. I imagine we'll see the court cases coming up soon. LoL Is the music you create available anywhere for listening?
Hey Jeff, good video. I think you'll get much better Audimee results if you would fine tune your own vocal performance (With Melodyne or Repitch etc.) before processing it through Audimee. This groundbreaking stuff for songwriters/ producers who can do a good/ great vocal performance but simply don't have a world class voice. Audimee will do wonders if you can get a really good vocal track of yourself and then take it to the next level. For female vocals, you can also pitch shift the tracks in your project (minus the drums) a few semi-tones (like 4-5) and then do the same with the vocals. (Or built your tracks in a higher key, then pitch shift it down temporarily to add your own vocal performance. There are many ways if you plan it carefully.)
Thanks! Still trying different techniques for best results.
I wrote a song that I felt like would be much better with a female vocal. I used Audimee to convert my vocal and I'm very happy with the result. Keep in mind, I started with a very clean vocal. The only processing I did with the original was a little compression and tuning. I did have to transpose the song because a female voice in the key that I sang it in just didn't sound good. I used the key I wrote the song in when recording my vocal and then changed the pitch to +3 in Audimee. The converted vocal had the same feeling and same emotion that my original did. It works great.
Excellent! I use it myself and get pretty good results.
I would really liked to have heard what audimee did to your vocal track with all the different voices , this will be way different from their somples.
@@AbrahamCloud I’ll create a link to some samples…..maybe SoundCloud.
Or you can use one good vocalist, and then use Audimee to generate all the background vocals.
Omg the auto tune is sooo obvious. I put a vocal with no music but the acapella. Maybe I'm not doing it right.
2038: music and creators are totally irrelevant. Subscribe to feel reality. 🤦🏻♂️
Maybe - but I don’t think so. Many AI song generation tools have no audience willing to pay. If you can’t monetize a service, it won’t survive.
just to clarify there is nothing free, independent artists and without conditions audimee is free only for feedback, ideas and training, it's just to try the free plan does not allow you to use the vocals to distribute your songs, it's good to read the terms before using or distributing your music, however I can say that it is on a good track and works well with small and not so serious errors!
True.
Great vid Jeff and thanks so much for sharing. As a musician who only needs some backing vocals on my songs, I think the free version suit me just fine. I'd just be uploading the chorus lines from my songs and picking a female singer. I'd never need the cover version section.
I just want to make my singing better. And add it to a popular song
What happens when you use up the entire limit - 15 minutes, in the free version? Will I have another 15 minutes next month? The reason for my question is that the word 'monthly' does not appear in the free version.
I’m not exactly sure, but after my free hours were used, I had to pay a monthly subscription.
So inspiring. I have been doing the same thing - I’m a songwriter with an ok voice and I’m enjoying using it. I have the pro plan and am trying different techniques to actually get a pro, usable vocal where the artifacts are not too bad. It depends on the model you use and how you record and articulate your words. I’d love to share tips snd tricks as I’m on a mission to actually get something professional quality - on rock stuff with a denser backdrop it is passable but at least with my voice I haven’t been able to use it on quite passages because of noise and artifacts. I input clean vocals from my studio. Anyway, thanks for the video! :)
Awesome!
This is the first video that I find that shows results on custom voices. Thank you very much. It would be nice to see a good workflow in order to get the best results when training vocals
Good idea!
I love this I got it too
Back up to the part you could sing ok lol seems like your skills have faded with age :)
For sure, sadly.
@@JeffyG I got it ! You were right it does amazing for people like me that really cant sing , do your best it will do the rest ! Thanks bro !
Just what the world needs....another subscription service....
Amazing! Really scared of the AI coming to music production! What do you think will it replace the Mix master engineers and producers?
Wow. Scary.
All AI tools are a little scary, but I find this useful. Remember, it's not generating the vocals. You still have to create a vocal track - but then you can pick a pro voice clone to take it to a higher level.
How many ai singers are there in total? TY! x
Looks like about 15.
@@JeffyG thank you! I purchased yesterday 🤪
@@RRADOMAL Cool! I’m using it and having fun.
How do you bounce in place or export mono vocals in Logic?
Right click the region of the vocals, Save As Audio file….I use the *.wav format.