Hi, I also couldn''t get a proper instrumentsl out of LoudMe - it always pit a small vocal section st the start. A strange AI generator, but shows some promise. Your spiece however is brilliant!
With this piece, I got the same feeling listening to the output for the first time as I had with some of my best MuseNet pieces, even though with those, of course, the composition process was more collaborative. And some of the other things I generated with LoudMe are pretty nice, too, although not like this, unfortunately. But like this, I will need to convert it to MIDI to work on them, and also to just obtain a better audio quality. As I mentioned in my comments below the video, I blew through all my free credits to use LoudMe on the first day of using it. It let me generate about 20 pieces before I hit my free limit, and the pieces are mostly between 2 and 4 minutes long. Udio, however, let's you work by training their model, to generate 10 free credits every 24 hours, and not only gives you 100 new credits a month, but also gives you 10 credits every day. If you don't use it for two days, you will not get 20 credits though. But later in the day, before I had used all my free credits with LoudMe, I think maybe I did something to increase my chances of it not putting vocals in the music even when I have "Instrumental" selected. Instead of starting the advanced interface by letting it "auto generate," I selected "instrumental" first and in the top lyrics box I put only the letter "a". I wish I had more free credits to experiment with what prompts do what. Before I subscribe to LoudMe, start paying the $8/month to use it more, I'd be interested in hearing the opinions of my listeners, what your experiences were like with Suno, Udio or LoudMe. Just recently, Udio let me produce a classical sounding vocal where the voice is excellent, and the piano quality satisfactory, and LoudMe won't do that. It has a kind of grit in the sound. However, considering the 20 generations I did with LoudMe, the musical and structural qualities of the pieces are more consistently good than those I've done in my few weeks on Udio. If anyone didn't read the comments below this, I spent hours fixing this piece in my MIDI software, as the audio-to-MIDI conversion I use is rarely perfect. People who are less inclined to go through what I did with this, removing the vocal spots, doing a conversion to MIDI, they'd be better off trying Udio.
@@Timzart7 As I mentioned in my earlier comment, LoudMe is using Suno to generate its music. I uploaded a couple of videos on my channel over the weekend showing this. I would be very cautious about paying for LoudMe. I use both Suno and Udio (and am one of the "guides" on Suno's discord server.) I find Suno gives pretty nice song structures, and works well for punchy/short instruments (plucked strings, percussion etc.) It generates up to 4 minutes to start with, but my personal preference is to keep extending from an earlier point as I like to change things up as the music progresses. Udio has really nice vocals, seems good for smoother sounds (bowed strings, synth pads, etc.) and has a wider sound overall. I haven't tried making percussion with it recently but it at least used to sound a bit watery for sharper sounds compared to Suno. The shorter generation length doesn't bother me much but I do like to be able to start listening almost straight away with Suno. Both have upload functionality so I sometimes take audio from one and upload it into the other which can help to refine the sound quality a little. I'm mainly interested in getting samples from them to build upon.
@@FurrybeansMusic Good to know, thanks. I watched both your videos on it at your suggestion. I read another caution about paying for LoudMe. And since I only used Suno for less than an hour, I'm going to have to try that more, especially if it is, as you suggest, as good as Loudly, because Loudly is using Suno. If I were not interested in mainly piano music, I'd be impressed with the sound quality of SOME of the Udio output, like a classical song I generated that sounds like Schubert, and the voice is beautiful. LoudMe, to me, all the output sounds fuzzy, or slightly buzzy in places. And the sound quality of the Udio French Pop piece I posted, "Sous le ciel de Paris," is quite nice in its raw state. I'm trying to generate piano things that sound like either Debussy and Ravel, and since these generators balk at suggestions you generate some music that sounds "like" a specific composer, and Udio rarely produces something when I put "Impressionism" in the prompt, that sounds even vaguely like that style, which is use of the whole tone scale combined with chromaticism, I guess I'm going to have to be more creative in my prompts. If you have any suggestions on the best free audio-to-midi converter, I'd be interested in trying some others. I just put what I thought was a simple Mozart-style piece into the one I use this morning and it really demolished it. It was too staccato or something, and it made the converter throw the rhythm off. I'm going to try quantizing it.
Hi, I also couldn''t get a proper instrumentsl out of LoudMe - it always pit a small vocal section st the start. A strange AI generator, but shows some promise. Your spiece however is brilliant!
With this piece, I got the same feeling listening to the output for the first time as I had with some of my best MuseNet pieces, even though with those, of course, the composition process was more collaborative. And some of the other things I generated with LoudMe are pretty nice, too, although not like this, unfortunately. But like this, I will need to convert it to MIDI to work on them, and also to just obtain a better audio quality.
As I mentioned in my comments below the video, I blew through all my free credits to use LoudMe on the first day of using it. It let me generate about 20 pieces before I hit my free limit, and the pieces are mostly between 2 and 4 minutes long.
Udio, however, let's you work by training their model, to generate 10 free credits every 24 hours, and not only gives you 100 new credits a month, but also gives you 10 credits every day. If you don't use it for two days, you will not get 20 credits though.
But later in the day, before I had used all my free credits with LoudMe, I think maybe I did something to increase my chances of it not putting vocals in the music even when I have "Instrumental" selected. Instead of starting the advanced interface by letting it "auto generate," I selected "instrumental" first and in the top lyrics box I put only the letter "a". I wish I had more free credits to experiment with what prompts do what.
Before I subscribe to LoudMe, start paying the $8/month to use it more, I'd be interested in hearing the opinions of my listeners, what your experiences were like with Suno, Udio or LoudMe. Just recently, Udio let me produce a classical sounding vocal where the voice is excellent, and the piano quality satisfactory, and LoudMe won't do that. It has a kind of grit in the sound.
However, considering the 20 generations I did with LoudMe, the musical and structural qualities of the pieces are more consistently good than those I've done in my few weeks on Udio.
If anyone didn't read the comments below this, I spent hours fixing this piece in my MIDI software, as the audio-to-MIDI conversion I use is rarely perfect. People who are less inclined to go through what I did with this, removing the vocal spots, doing a conversion to MIDI, they'd be better off trying Udio.
@@Timzart7 As I mentioned in my earlier comment, LoudMe is using Suno to generate its music. I uploaded a couple of videos on my channel over the weekend showing this. I would be very cautious about paying for LoudMe.
I use both Suno and Udio (and am one of the "guides" on Suno's discord server.) I find Suno gives pretty nice song structures, and works well for punchy/short instruments (plucked strings, percussion etc.) It generates up to 4 minutes to start with, but my personal preference is to keep extending from an earlier point as I like to change things up as the music progresses.
Udio has really nice vocals, seems good for smoother sounds (bowed strings, synth pads, etc.) and has a wider sound overall. I haven't tried making percussion with it recently but it at least used to sound a bit watery for sharper sounds compared to Suno. The shorter generation length doesn't bother me much but I do like to be able to start listening almost straight away with Suno.
Both have upload functionality so I sometimes take audio from one and upload it into the other which can help to refine the sound quality a little. I'm mainly interested in getting samples from them to build upon.
@@FurrybeansMusic Good to know, thanks. I watched both your videos on it at your suggestion. I read another caution about paying for LoudMe. And since I only used Suno for less than an hour, I'm going to have to try that more, especially if it is, as you suggest, as good as Loudly, because Loudly is using Suno.
If I were not interested in mainly piano music, I'd be impressed with the sound quality of SOME of the Udio output, like a classical song I generated that sounds like Schubert, and the voice is beautiful. LoudMe, to me, all the output sounds fuzzy, or slightly buzzy in places. And the sound quality of the Udio French Pop piece I posted, "Sous le ciel de Paris," is quite nice in its raw state.
I'm trying to generate piano things that sound like either Debussy and Ravel, and since these generators balk at suggestions you generate some music that sounds "like" a specific composer, and Udio rarely produces something when I put "Impressionism" in the prompt, that sounds even vaguely like that style, which is use of the whole tone scale combined with chromaticism, I guess I'm going to have to be more creative in my prompts.
If you have any suggestions on the best free audio-to-midi converter, I'd be interested in trying some others. I just put what I thought was a simple Mozart-style piece into the one I use this morning and it really demolished it. It was too staccato or something, and it made the converter throw the rhythm off. I'm going to try quantizing it.