Suno Audio Extend! How does it compare with Udio?
ฝัง
- เผยแพร่เมื่อ 26 ก.ย. 2024
- Suno Audio Extend! How does it compare with Udio? suno.ai
To follow up our recent video featuring Udio’s ability to extend your own audio, today we feature Suno’s similar capability.
While they both get great results in my opinion, there are some fundamental differences between how the two deal with the audio you upload.
Today, we’ll run the same audio snippets through Suno that we did with Udio and see what we come up with.
To check out the results with Udio, watch this video:
• Udio UPDATE! Extend yo...
There is a ton of potential with both of these tools for creative inspiration and getting rid of musicians “writers block.”
👍 LIKE If you found this video valuable. 🙂
🥰 SHARE If you know someone who might enjoy this video.
⏬ DOWNLOAD or ADD This video to your PLAYLIST for easy access later.
💬 COMMENT Your thoughts and questions are welcome!
📝 SUBSCRIBE / @bobdoylemedia
That's what keeps me going!
🔔 And make sure to hit the NOTIFICATION BELL to stay updated! 🔔
🌎EXPLORE
bobdoylemedia.com
🗓️ MY HISTORY
🧠 meetbobdoyle.com
🤫 Featured Law Of Attraction and Neuroplasticity expert in the book and film, “The Secret”. • The Secret
🎙️ Voice Over Artist 30+ years
📻 Broadcaster / Actor / Creative
❓ASK ME
Got a Question? bobdoyle@me.com
📌 FOLLOW ME
Facebook: / bobdoylemedia
Instagram: / bobdoylemedia
🪙SUPPORT
If you want to support me, the best thing to do is to share the content… sharing is caring!
If you want to support me financially (completely optional and voluntary, but a lot of people have asked for this):
www.thebobdoyl...
Learn more from Bob Doyle Media: www.bobdoylemed...
It’s a shame Suno’s audio is such bad quality, because it’s arguably more creative. They just need to add multi-directionality and create Udio level audio, and it would be my default.
More coherent, yes, but more creative? Not sure about that.
Hope 4.0 has massive audio quality improvement.
I think the Udio actually samples real singers... that's my guess as to why the singers sound better, it's because they are using real voice samples.
They put in 90's polka singer... and they got Weird Al. They didn't specify his name.. it generated the song with his voice.
nah, udio is more creative
I assume you're referring to vocal audio quality being subpar? In my experience, udio doesn't sound better. Instruments don't always come through clearly.
Suno needs to upgrade their MP3 compression from 192 kbps to 320 kbps like Udio.
Yeah. You right
Suno does not know how to mix and master tracks yet, then need a full 3 db reduction on the entire database, these MIT kids dont have experience like us pro's need. But, this is a tool of the future.
I've been "playing" with SUNO almost since its inception. What I've personally noticed (and I've made hundreds of songs) is a change in voice quality from version to version. In its first version I was speechless at how natural the voice sounded. Today every song I hear from SUNO , no matter what kind of music I ask it to create, is accompanied by a voice that is ALWAYS distinctly synthetic. It almost always sounds like the singer has a trap vocal filter on. The orchestral part of SUNO has been SIGNIFICANTLY improved. Now the music sounds much better, clearer, with more clarity. In seconds you have a ready song. NOT usable for anything more than just listening to it and enjoying it. On the other hand, UDIO will take your soul out of your feet BUT the result, with a little mastering, is ready to go as a demo to some record label. You will spend hours, maybe even DAYS to complete a single song. But honestly the end result is Unquestionably BETTER than that of SUNO. Many times it is IMPOSSIBLE to distinguish that it is a product of artificial intelligence (UDIO). I recently wrote a song about some personal experience (music and lyrics). I gave the music as a "base" to UDIO so that it has an idea of what I want and then the lyrics. The end result honestly gives me goosebumps and tears (ok I don't expect anyone who listens to this to cry, I relate to the lyrics). Interacting with UDIO is an intense, long, EXPERIENCE. While interaction with SUNO is something of a "one-night stand", fast, beautiful but not (usually) sustainable.
Merci beaucoup pour votre commentaire très pertinent. Je commence à utiliser Udio mais j’ai bcp de mal avec les 38 secondes et rassembler les morceaux. J’ai l’impression qu’il faut donner énormément de crédits pour tomber sur une chanson correcte. C’est un peu de la loterie non ? Avez-vous des astuces pour que cela devienne moins aléatoire ? Merci à vous et le meilleur dans vos projets musicaux !
@@brunogama8537 so i had the opposite experience (can also just be a matter of taste), i entered the same prompts and lyrics in udio and suno and where suno gave me directly what i wanted, udio deviated miles away from what i had entered, it didn't use the instruments i had requested in contrast to suno. i admit udio can make good music but in direct comparison it didn't convince me personally.
you can extend the songs by another 33 seconds, but it's a matter of luck whether the song remains good. i had tested that too and for udio it was over after 3 - 4 minutes, after that there was only anything but great music. i often had to repeat before i had something to continue with. as i said, it may be that others were luckier than me and that udio is not for me in terms of taste.
Very nice feature. Good that we have a competition. I hope that Suno eliminate finally this "robotic" voice weird echo and then... will be even more interesting ^^
If suno had the great voices of udio, and you had the opportunity to get the download as multichannel with each instrument separated, to work on more yourself.. it would be amazing
I like Suno much more for modern airplay songs
Still enjoy UDIO far more. UDIO gives me giant operas some of them hours long. They sound exactly like Rossini, Donizetti, Mozart, Verdi....Blows my mind. I don't see me ever leaving Udio.
Very true. Udio is amazing for all those reasons!
I don't know why people are saying Udio better, in my opinion Suno is five times better in sound, speed and clarity than Udio.
It's true. Udio is far away more impressive but the only problem is that you needs to generate many extensions, in my case, and I am a subscriber standard, Udio generates always no matter what I do, 30 seconds, even if the option 2 minutes is on. It could be great, if Udio add the full song option in the first statement. . Anyway, Suno and Udio are great, I created many songs for an AI influencer, I need Sora or kling or luma labs to see her alive. Emote would be better than Sora. 😂
In SUNO, unlike UDIO, to receive the entire song, you have to click on the three dots and ask for the complete song, so that you will receive the exact section you uploaded at the beginning of the song
I worked with a (pop) songwriter in the 80s and his preferred method was to listen to a short section of music and then tag on what he thought should follow it. He would then bin the inspirational section and, hey presto, a derivative work that no one could tell was derived. This AI feature is not so different except that it spreads the inspirational part through the derivative piece.
Thanks for another really interesting video. The extension features definitely give both Udio and Suno new relevance for music creation. They both work great in their own way although I do prefer how Udio adheres to the initial upload in a much more strict way compared to Suno. Also just have to add that you're definitely my favorite creator here on TH-cam, your reactions to some of the more bizarre creations are priceless. Thanks so much for sharing your knowledge and creativity, and for your hard work in making these videos.
It extends the original audio file you sent, so you can generate the full song (like with regular extends), and then you'll get your audio and then the extension.
When I tried to extend from the middle of a piece, it continued from there and was very good in keeping the structure and melody of what came before it in future verses/choruses.
However, I like the idea of uploading a piece with silence at the end, and then allowing the extend to generate a song that starts with some kind of intro and incorporates your piece.
That he generation at around 25:00 was stunning!
Wow! Thank you. it was touching! Did you cry after 20:08. I would have.
I found that most of my music I uploaded it kept it true to what it was a few of them they changed completely
I don't think Suno is changing your clip. When you upload a clip to Suno and ask to extend, you'll get just the extension - and only then you can ask to get the whole song, which will be the merge.
That's how Suno always worked. Udio always automatically gives you the merged version. Both approaches have their merits - I kind of like the Suno way but the Udio approach is simpler and lets me judge the piece more directly.
Composition-wise I prefer Suno, but the technical quality of Udio is indeed better.
You should read the copywrite info from suno and ubio/ You will never get copywrite are own it. and you can be sued for using it.....Buyer Beware
This is the point I’m trying to make. Only use it for inspiration or non-commercial use until this all settles down. There are plenty of legitimate ways to use these tools and create music that doesn’t infringe copyright.
@@BobDoyleMedia And GOOD is that! You actually still have to be talented and put up some god damn effort in order to make money with music, as it god damn should be.
Yes I started snapping with you, and stopped right before you said “keep snapping”
I use both, in udio i create the base and in suno i make long song, the i use udio yo make the intro i like it
Same, but I'm doing the opposite way. First Suno then Udio to modify parts that I don't like it, and to create a more interesting intro and outro. I wrote Udio days ago, about to include FX sounds, like people talking, fireworks, crowd in a stadium, because Udio generates all of this but randomly, and they replied me : that thing is in our roadmap "😮
I do that too😁
Please click on the 3 dots after extending and choose get the full song to get the song that included your part.
It’s not writing over it - it just for some reason only gives you the extension to the song and leaving out what it extends from. When you then select get full song you actually understand what it is doing. It’s not recreating or writing over your song. It’s extending from the end of your snippet.
Wow. Well that is a basic function I missed but could be a lot more intuitive! But thanks!
@@BobDoyleMedia yeah only figured that out after struggling for a while
All the hidden prompts it supports also only get clear after analyzing other people’s songs.
@@MaximilianFeichtinger "Hidden prompts"?
Sounds like you need a follow up. This is covered in their docs and makes sense after you know about it.
That Pony Dreams is marvelous!!
I use both daily. Udio is clearly stronger for audio and voices, however it is not as good at maintaining the coherence of the music structure if you choose the 2 minute custom mode. So for now, I find that Udio is clearly at a superior level in terms of audio quality
Great feature. I decided to see what would happen if I uploaded a small passage of text to speech. The speech was the first line or two from the song and was spoken. Suno created a song from my lyrics and the singer sounded very much like the voice I uploaded... slightly different, but people don't really sing 100% the way they speak. Of course, now that they have the longer generation length you can either complete the song without having to incorporate the talking part, or, if you want you can use it as a cool intro.
I commented on your last video, but you must have missed it. Suno works differently than audio As it doesn’t attach extensions automatically. In order to get the full song with your original intro, you must go to the extended songs menu and click the generate full track button
Just tried it recently . If you have clear ideas not random ones , that feature will helping you to check or test vocal line ideas . I believe it will useful for anyone who mostly stuck in 8 or 16 bar loops
wow, wow, wow. I uploaded my separated vocal and it even sounds similiar to me.
Right! They’re clearly doing a bit of cloning!
I like Suno's creativity and catchiness more than Udio's. Whoever gives us all separated stems next, will win the next big was of customers. With all stems to our disposal, we can increase the quality of each instrument by using Synthesizer V or other plugins and compensate the lower quality sound Suno has nowadays compared to Udio.
Hope they add in painting as well as separate tracks for the vocals and various instruments.
I said this months ago on discord and someone tells Mex what you talking about Willys? 😂😂. Then Udio add inpainting. Kneel before more mere mortals! 😂😂😂
Really love that "Easy" track. Would love to hear the full song.
What are you using to create the little animations for the songs?
19:11 this one is so beautiful. I would love to hear the full version of that.
Amazing content! Thank you! I made quite a few songs with Suno and love it. Haven't tried udio so cannot compare. Thanks again for the videos on AI for music. :)
Thanks for watching!
I was snapping along at home :)
“Bryan Mills” at the end! 😅
For the moment : Suno is FASTER but not the best quality. Udio is way slower (for a full song) BUT the quality (overall, quality, versatility , final result etc) is light yrs in front of Udio.(my opinion)
Now gotta try it
Suno's hit or miss for me. When I get an idea in my head for a certain style of song, I kinda want it close to how I envision it. Sometimes the music is spot on but the vocals or cadence is off....or vice versa. The "add audio" feature is pretty cool though. I uploaded a a quirky jazz guitar line that I recorded years ago, and I was pretty blown away how well the song was structured and I only had to listen to the second sample. I wanted a drums, bass and guitar fusion piece and it nailed exactly what I wish i had the talent to make from the beginning. lol
Suno is my bread and butter. The audio upload feature is bitter sweet however. First reason being the longer sample you use reference, The longer sample is stuck in it, for better or for worse.
This is unlike Udio, where once you do your audio upload, you can go ahead and work backwards to overwrite the uploaded clip, which is great if you wanted to add words to a sample that had no words.
Secondly, I really, really, really hope they can solve the background noise degradation issues with V4.
I'm doing my best to work within it's limits, but it's heartbreaking to walk away from a 80% finished track because the increasing distortion absolutely ruins the vibe.
Not very good for ambient music, which is what I happen to create with it.
Lovely Bob! How do you create the video to your songs??? plz share, god bless
I hope there will soon be an option to enter a melody, for example for a chorus. This way I could reproduce my old songs in different ways
Great video Bob! I don't have the money rn to get either subscription, but getting to see where these apps are going lately is awesome.
Also, I'm not a big country music fan, but ever since the Udio video I've had "I went down to the farmers market today" stuck in my head, it's too catchy lol
awesome sir
I definitely like Suno's extended version better than Udio.
What do you like about it better? Just curious.
Good thing about udio extend is adding intro.
@BobDoyleMedia I like it loads really fast. And the prompts are consistent.
Suno generates absolutely plastic music. It sounds like 96 kb/sec midi. Udio is much much more fancy and diverse. Currently, the main disadvantage of Udio is that it can only be generated for 30 seconds at a time
@@Kutiepov they’ve extended that to 2 minutes.
I think AI invited the ghost of Leonard Cohen on one of the klezmer song. ;-)
I feel Suno puts out better songs, both Udio and Suno are great I can only imagine where they will be in another year or two, quality and more control should be through the roof the speed they are improving things.
I've tested with my old own song, but It have cloned all the noise and bad quality too :D
Bob i have some news directly from Udio. They'll include in the next updates: FX SOUNDS CONTROLS: a mean if you need a crowd, need some sounds like epidemic sound, well, you'll have the control over those effects. I wrote Udio and they replied me: " that thing is on our roadmap".
Great job you do with all your videos.
you should share your creation in suno
I also noticed that when I uploaded a track with lyrics, and I incorporated those lyrics into the lyrics I'm sending to the Extend - it would incorporate the relevant parts in the extended song!
specifically, I uploaded some scatting, and when I wrote the specific wording of those scats (e.g., tum-taka-taka-taka-tum-taka-tum), it would insert that part from my audio into the song in the position I wrote it in the lyrics!
@@DjNaSh Yes this is a hack which allows one to get a better version of what they uploaded.
Hello Bob! Thank you for your videos, I love them and I really learned a lot from them. I own a dance school in Germany and generate songs for customers who have birthdays and special celebrations. Now I'm wondering how you got the software to generate a slow 3/4 beat. For me this only happens every thirtieth attempt. :D Thank you for your tip!
What program do you use to create the animations in the video that play above your head as you listen to the songs?
Wowwww at 2:30 🤯❤🎶🙌🏻⚡️💯🪄
Uncle Eddy is low key a fresh vocal and drum recording away from sync licensing
I have Udio and as a poet and electronic musician I love it…would not like Suno rewriting my lyrics… do like you can see the generated lyrics before commiting them to a piece… audio quality does not sound as good as Udio… how are credits spent? thanks for the demo… as a Man in his mid 60’s I love and hate this technology… but it is still in its infancy
Sorry if this as been mentioned before but can you create an Acapella with prompts instead of the full track. Many thanks in advance
Fingersnaps at 12:30-ish - something about your recording setup inserts a delay on your voiceover compared to the Suno playback. . . (or my rhythm sense way off).
No, you’re correct. For whatever reason the timing got way off in that section of the edits and we had to try to manually fix it.
Speech-to-speech too? Or text-to-speech only?
Udio is best
Both are best anyway 😅
Sadly neither can “ add new instrument layer” this would be a game changer. Say you upload a 40 second section of a track and add “90s sounding pad synth in d minor” this would change everything ! They need to add this feature and then let us download the new stems and build the tracks !
I don't know why my last comment as been deleted or something, but I'll try again without sharing my 2 tests.
I really enjoy Udio more than Suno. Suno (3.5 at least) struggle big time to create more heavy genre (psych-rock, power metal, etc.). The vocalist always sounds like an autotuned popstar. I tested using the same audio snippet, prompt and lyrics in both platform, and while it took me hours to finish my song in Udio, I'm much more satisfied with the result than anything Suno gave me. And I don't even talk about audio quality here... Udio sounds way better than Suno for now (IMO).
Its called shadow banned
Everything we say or do on any platform, facebook, youtube, X, ect is foot printed. If you fail the Algorithm on any platform it follows you. Kinda like a spreadsheet the FBI uses on individuals. So be careful what you say, do, write, eat .. understand
Creativity > sound quality.
Use it for inspiration, not just generate music and blatantly claim it as your own.
thanks a lot for all your great tutorials on these wonderful AI tools. I have a specific problem today though: how can I change some words of the text from a otherwise finished Suno creation in a way that everything else stays the same? I only see the option to create a whole new song. I guess I have to upgrade in order to be able to change a detail like this, right?
I found that the max time duration for the extended part is up to 2 minutes, but it looks like got it even longer. How did you get that?
The real question is which one give you full rights to use the songs and monetize them.
None of them. They are trained on copyrighted material, and the EU recently passed a law saying you can't generate revenue with AI music without crditing and compensating the people who trained it with their work (millions of ACTUAL composers and musicians). No, for monitization i'm afraid you need talent and effort and make your own god damn songs as opposed to asking something else to make it for you.
You should use a more popular music style -
I made a great song with Suno Ai but the beginning is so bad..I just want variations of the beginning without messing up the whole song in the same format..There is no explanatory video on this subject. I want the same song in full instead of extending the song...what should I do _?
The compositions are nice, but the singing is really not OK. I uploaded a fragment of an audio cassette. The audio cassette had more dynamics than what Sumo generated.
Hi bob, how bout uploading a melody line with a chord progression and see what happens?
Very Randy Thumbnail
Admittedly, I am mostly on an ancient computer when working with Udio and Suno, both, but while I can easily upload my audio to Udio, Suno takes so long that I have yet to not cancel because I have no idea when the upload will finish. I have a fairly newer laptop, and haven't yet tried to see how the uploads work on it. My desktop (the ancient one) is far more conveniently located for me, which is why I use it more often.
I laughed at the typical AI rhyming part. I agree, AI uses the laziest rhymes. I have better luck when telling the AI to not use a rhyming scheme and/or asking for it to write in blank verse. If it gets to desperate measures you can tell it that all the rhymes are copy-written lol. It still makes rhymes but they are less lazy.
Can I ask how you got those vivid moving video done when you played the song? which program is easier to generate for the videos? thank you. I use the SUNO for two weeks already.
You're giving them your data! What are you doing?
😄
Suno is much better. Udio extends only 32 sec, too short.
the a.i is being generous... lol
how do I define key and tempo?
Does anyone know a [prompt] to make suno come to an end fast? I have the problem, that it goes on and on and even starts again with a variation after the song has already ended. Enough is enough. [end in 15 seconds] works sometimes but more often it does not and even sings "End in 15 seconds".
Can Suno extend infront of the clip so it has a nice intro?
No
@@Fheym ah bummer, thank you for letting me know.
@@Melodic_Analysis Actually, you could try extending the clip, and then in the custom lyrics section use the [Intro] prompt and then set the start time at 0:01 and that should generate an intro for you. That's the only workaround I can think of that might work.
@@Fheym oh that's an awesome idea! I'll have to try it. Thank you!
Hey Bob what if you upload a 10 second random acapella , followed by an random different guitar chord of also 10 seconds , followed by a random beat in a different bpm. All these 3 different random sounds in one self put together wave file of 30 seconds in total. Would sumo be able to create something decent of that experiment into 1 credible new song ? Let me know haha interesting. Because it won’t yet let you upload and combine several uploaded wavs. I feel this is going to be a next step to update in the competition race.
AI animation is still really weird (It looks like a nightmare) 😂
But the music side, like Suno, has evolved a lot
Think you need to extend a few seconds before the end of the clip.
Is it possible to use my own vocals in any of them?
25:25 "I will find you, and I..." kill you? OMG :)
Can you please tell the AI tool you use for the video shorts? (pay site are ok to me) I'm asking because I spent a lot of time, trying a number of them, but only discovering they are all immature and deforming the images well before the 4 seconds,
Hi, do you know how to change the names of the songs produced in Udio? Thanks!
When the song is on screen, you should be able to click an "Edit" option, where you can change the title and pictures.
@@BobDoyleMedia Perfect! Thx! =)
yOOOOO00O0O0O0O00O0O00O0O0O0O !!!
oh my god
Say it man, Suno's on top again in a way!😂
Where did you get that video from?
I’m not sure what video you’re talking about. My editor often generates video using Leonardo.Ai to complement the music generated by the music program. Is that what you’re talking about?
@@BobDoyleMediathe video that plays along with each song demo! 😭 the video clip that a ton of people have asked about in the comment section that have been left unanswered before this response!…
your emotions ❤
With Suno, you should never put artist names in the genre box. It was purposely trained without any artist names, so it literally has no idea what those names mean, and it just confuses the AI.
to be honest, it ia a bit silly to pur in something so basic and here you go....... a nice piece of music is created. AI puts out the music, which ia obviously already there to some extent, and which is somewhat a bit related to the input. Just dont see a big chunk of creativity from both sides.
Suno is awesome but quality is horrible which makes it unusable
For me it doesn't take what I've uploaded and incorporate it in any meaningful way or at least in a way that I can recognize.
I've had that experience too. It's kind of hit or miss sometimes. How long is your sample?
38 seconds.
Yes, hit or miss is a better way of describing it actually. But I like Suno better than Udio, anyhow.
suno is better at making what you type from scratch but is so much less creative. It always makes it sound mainstream. Udio has far superior vocals while Sunos vocals are really bad.
funnyyyy :)
11.45 you extended from the wrong snippet.
Yes, I finally caught what I was doing. It happened to me a couple of times.
We've all done it lol.
If you click to get the "Full Song", at least for the times I've done it, I've found that the original uploaded clip ends up being reconstructed by the engine. It's not "exactly" the original audio coming back.
I also tested this and uploaded a complete finished music production from 1990 (own production). A completely useless result. Horrible quality and absolutely unusable. Same with UDIO. Ultimately, the order is just reversed, echoes are added and the melody is shifted and layered on top of each other. Catastrophe.
Зачем рот так открыл на картинке? Неужели не понимаешь, что глупо получается?
wow you're old
Tell me about it...
udio way better