General feedback

I use UDIO extensively for professional music and vocal production. When UDIO first entered the market, I was SHOCKED by its incredible quality, and from the very first days, I paid for the PRO subscription. Then updates started rolling in, and I began to see complaints, but I didn’t pay much attention to them because I was still able to achieve excellent results. I didn’t complain because I clearly saw the high level of what I could create with UDIO. As it happened, I didn’t use UDIO for a week or two. When I returned and started working on music again, I felt like there were issues. I applied all the skills I had in using UDIO, but I couldn’t achieve anything I wanted. The music sounds weak, the vocals are muddled or robotic... everything is really bad. Then I went to listen to what’s on the front page of the UDIO website and heard the same problems in all the new tracks by users. In shock, I went to Reddit and found a review that matched my experience. Then I came here and saw similar feedback here as well. Yes... I can confirm—things have gotten bad. What I loved so much about UDIO has almost disappeared. This is unacceptable. Unfortunately... I really want an honest and detailed response from the company’s leadership. But I guess that’s impossible. I really wish I could say, "It seemed that way to me," but I’ve worked too much and for too long with neural networks in general and with UDIO in particular. Unfortunately, it’s not just my imagination... the good old UDIO, as we knew it, is no longer with us in the form that was so appealing to true connoisseurs of quality. Its one competitive advantage is gone... it no longer exists.

Something terrible has happened with prompt > output results in the last month or so. My go-to prompts that would reliably provide amazing sounding results are now no longer doing so. I have confirmed this by testing several different tracks generated before this regression, using the exact same prompt, Model, and seed#, and they generate results that are close but sound off, or completely not the same as what they used to be. Both models are affected. You can view this playlist to see the comparison: https://www.udio.com/playlists/ojTci1TsXrkr4GceaQmqXn This leads me to believe that something has changed with how the prompts are being interpreted to the actual results. It takes AGES to land on prompts we like, and now it seems I have to start all over again, or (if the models are permanently nerfed in some way) abandon Udio altogether. Something has definitely changed and I am not happy. I have already canceled my yearly subscription renewal and will not renew it unless I can find some prompts that sound good or the old prompt structure works once again.

Nervend lauter Gesang, die Instrumente werden "verschluckt", also hinter dem Gesang runter geregelt oder verschwinden ganz. Die Songs sind immer oft nach dem gleichen Muster aufgebaut, dabei nerven absurde Intros, Outros und Gitarrensoli mit schrillen Geräuschen. Bei Verlängerungen ändert sich oft das Thema, die Instrumentierung und der Gesang. Remixe werfe ich zu Hauf in den Müll, man kann sich schon fast drauf verlassen, dass sie schlechter sind als das Original. Der Spass geht verloren. Ich hatte die Hoffnung, die KI lernt dazu und wird besser, bei Udio ist leider das Gegenteil der Fall. Ich werde mein Abo wohl kündigen.

Hello, Udio Team, I have been consistently using Udio for nearly a year to create music. Since version 1.5, I have been highly satisfied with Udio's performance, to the point where I believe it has far surpassed SUNO. It has even produced Korean-language songs of commercial album quality, showcasing its impressive capabilities and expanding creative possibilities. However, I have recently noticed significant changes in Udio’s internal processing. These changes feel like more than just a simple update—they seem to have fundamentally altered the way music is generated. The issue is that these changes appear to have negatively impacted the overall quality of the music. Changes in Mastering Processing Previously, each song had a distinct character, and the mastering adapted to the mood of the track. However, now the mastering seems overly uniform, making all songs sound similar. As a result, individual songs lose their unique identity, and the diversity of musical expression feels restricted. Changes in Vocal Equalization The vocal mixing has also changed. Compared to before, the vocals now sound less natural and sometimes seem to have EQ processing that does not match the song’s atmosphere. Previously, the vocal tone and emotions were better preserved, but now they often feel overly standardized. Loss of Variety When Combining Multiple Genres In the past, when I included multiple genres in the prompt, the generated music reflected a mix of styles with different characteristics. However, recently, even when I input multiple genres, the output tends to have just one fixed characteristic, lacking the variation and uniqueness that was previously present. This makes the sound feel more constrained and less dynamic. Outputs No Longer Reflect My Refined Prompts For almost a year, I have refined my prompts through continuous effort to achieve a specific style and quality. However, the recent changes now produce output that is entirely different from what I have been working towards. This shift disregards the nuances and customizations I have developed over time, making it much harder to achieve the desired results. As someone who has created more songs on Udio than most users, I can confidently say that the overall quality of the music has declined since these recent changes. If these adjustments were intentional, I believe it would be beneficial to offer an option to retain the previous version’s processing (such as keeping the 1.5 version’s characteristics) alongside the current system. I sincerely hope that Udio continues to improve, and I would appreciate it if user feedback on these recent changes is actively considered. Thank you, and have a great day.

I have a big suggestion for you!!! why you don't remove the prompt field, the advanced options and the create button and instead of it leave only one button with "Lottery, try to win", cause really the model ignore completely the prompt , style reductions and most of other options and generate what he want

After hundreds of generations with the same prompts and settings UDIO is now in the past week forcing my generations to have spanish lyrics and or poor quality. Feels like something has changed with how prompts are processed in the past week or so. Something is wrong. Also if it's not Spanish then it's also hundreds of "in the shadow" lyric gens.

the voices in all my songs are getting worse and they USED to be clear... why?

I see a lot of negative feedback here. Ive created over 60 songs which I think are brilliant and I couldnt have done it without Udio. Yes, some things are more difficult to navigate now, but thats offset by incredible dynamics and more complexity within musical choreography. With complexity comes challenge to synthesise into simplicity and when it comes to the infinite infinity that is musical potentiality, its a miracle that we can even do what we can currently do - given the AI is working off code rather than sound itself in order to generate sound that we enjoy. I think - be grateful. Trust the process. I for one am fucking in love with the fact that I can co-create music with AI that matches my soul resonance and share that with the world. Yes, other platforms may exist and yes they may have improved in ways Udio hasnt yet (maybe) but Im sure a lot of cross-referencing is going on between these platforms and ultimately they will evolve together when one sees whats working and another is like, hey this could be better. Thanks team, for everything.