Spotify AI covers and remixes: licensed tool vs an original score
Spotify and UMG announced a paid add-on for fan covers and remixes. A licensed remix tool is not an original video score. Here is the Sume route for the second.

Can you make AI covers and remixes of real songs on Spotify? Spotify's newsroom post of May 21, 2026 (read 2026-10-04) says it signed recorded music and publishing licensing agreements with Universal Music Group so it can launch a tool for fans to create covers and remixes of songs from participating artists and songwriters. It describes the tool as a paid add-on for Premium subscribers, powered by generative AI, with artists and songwriters sharing in the revenue.
That is a remix tool inside Spotify, built on licensed catalogs. It is not a way to get a file for your own video, and it is not an original score. If you need music that is new for a video, the route is a generator that takes a prompt and returns a file, which is what Sume's Music Router does.
What the Spotify post covers
Read the announcement on Spotify's newsroom. The points I verified there are the licensing deals with one major label group, a paid add-on for Premium users, generative AI as the engine, and revenue sharing for artists and songwriters who take part. I did not verify a launch date or whether the output can be exported, so I do not state either.
Remix tool or original score: the practical difference
A licensed remix tool starts from an existing song and returns a variation of it inside a platform. An original score starts from a brief and returns new audio you can place anywhere you have the file. For video work the second is usually what an editor needs: a bed of a set length, no vocals, matched to a cut.
| Question | Spotify covers and remixes | Sume Music Router |
|---|---|---|
| Starting point | An existing song from a participating artist | A text prompt, optional image_url |
| Output | Not specified in the post I read | Audio artifact on media.sume.com |
| Length control | Not specified | Steer in the prompt; duration is rejected |
| Vocals | Not specified | Say Instrumental, no vocals. in the prompt |
Writing an original brief
The Music 1.0 docs say the request is text first and recommend a scene-specific brief: emotion, genre lineage, tempo as a number, key and mode, two to four textured instruments, an arc with one named moment, and era. They add that these are creative directions, not guaranteed output settings, so listen to the result.
import os
import requests
r = requests.post(
"https://api.sume.com/v1/music-router/generate",
headers={
"Authorization": "Bearer " + os.environ["SUME_API_KEY"],
"Idempotency-Key": "remix-vs-original-001",
},
json={
"model": "sume/music-auto",
"prompt": "Proud and nostalgic synthwave, 108 BPM, E minor. Analog bass, gated snare, a bright lead that enters at 0:08. Full return at 0:24. 2024 hyper-clean. A 30-second track. Instrumental, no vocals.",
"mode": "sync",
"wait_timeout_seconds": 30,
},
timeout=60,
)
print(r.status_code)
print(r.json())What to do with the file
Use the media.sume.com audio URL as the soundtrack on a Timeline 1.0 render, or save it as the spine. The Music Router returns new audio, so there is no source song to clear; the licensing question for your own use is one to read from Sume's terms, which this post does not summarize.
Sources
Related posts
More in Comparisons
- Stable Diffusion alternatives in 2026: open weights or hosted API
SD 3.5 is still Stability's newest flagship image model. FLUX.2 [klein], Qwen-Image 2.0 and Z-Image Turbo are the open options. How to pick between them.
- Subtitles for a recorded video: 100 ms partials or final text?
MAI-Transcribe-2-Streaming sells 100 ms partials for live subtitling. For a recorded video, final text and word timings matter more. Choosing with Sume STT.
- Suno Premier's 30-minute upload versus Sume's audio limits
Suno Premier allows audio uploads up to 30 minutes. Sume's Timeline audio joins up to 1,800 s (30 min), while audio detach takes 1,800 s in and 900 s out.
- Suno Speech beta: voice and music in one pass, or separate tracks?
Suno's Speech beta makes voice and music in one track. Its blog lists wandering accents and long pauses. When to prefer separate TTS, music and a timeline mix.
Written by Sume