Genx Notes

Microblog Category: AI

  • The next day after I said that, I was making music with AI. Haha..

  • I’m making music again without AI. It’s satisfying.

  • By using AI to create beats, I realized how unique my beats were.

  • I’ve been trying out the remix feature of Udio 1.5 too much lately and I’m getting tired of it.

  • I think Udio-130 is pretty good compared to the previous 32-second one. However, I wish the duration was about 4 minutes like Suno.

  • Since I updated to Udio 1.5, vocals have often been included even when I output Instrumental. I’ve looked into various ways to remove vocals, and it seems that setting Lyrics Strength to 0% is the best.

  • There is a bug that if you make a 2:11 remix with Udio-130 and try to extend the beat, a 1:05 file is generated. If you refresh the browser and then extend, it will become 2:43.

  • Honestly, I think Udio has changed a lot since 1.5. The AI level has improved.

  • Udio 1.5 Update: A New Era in Music

    Udio 1.5 Update: A New Era in Music Production

  • Udio has been updated to 1.5.

    1. You can now download stems. Drums, bass, upper parts, and vocals. This is big. This is better than Suno, where you can only download instrumentals and vocals.
    2. Remix can now be done on tracks longer than 32 seconds. This is also a big deal. You can play around with the parameters to set how close or far you want the remix to be from the original song.
    3. Udio-130 (which generates 130-second clips with one prompt) is now available for Standard plan users, which is a nice change from the previous version, where you could only get clips every 32 seconds.
    4. You can now edit Clarity. It’s not just about separation, you can make it sound old-fashioned by intentionally making it worse, or vice versa.
    5. The sound quality has improved. This is great in itself. The “crushed snare drum” sound that was once instantly recognizable as a Udio is no longer as noticeable.
    6. This is not a new feature, but it has been talked about on Reddit for some time that Ultra is not necessarily the best Generation Quality. Some people say that Good is better than Ultra, and others say that a lower setting sounds better. It seems to be the same in the 1.5 version. It seems that you can experiment with various things just by adjusting Clarity and Generation Quality.