Does Lyria 3.5 music carry a SynthID watermark?
Google says all Lyria 3.5 audio includes a SynthID watermark. What that means for tracks made through an API, and what Sume's docs do and do not say.

Yes, per Google: its Lyria page says all generated audio includes a SynthID audio watermark, imperceptible to the human ear, and the Lyria 3.5 model card lists SynthID watermarking among its safety measures. Sume's docs do not describe watermarking on their own music pages, so treat Google's statement as the source and Sume's route as the way you reach the same model.
Google's statements are from its Lyria page and model card; Sume's from the Music Router docs. Read 2026-09-29.
What does the watermark do to the sound?
Google describes it as imperceptible to the ear and meant for identification. Neither page read for this post gives technical detail about how it survives editing, trimming or re-encoding, so do not assume it does.
Does Sume's Music Router add or remove it?
Sume's docs describe the router as passing your prompt to the engine and returning a Sume-hosted audio artifact. They make no statement about the watermark, so the docs cannot confirm it either way for a Sume-generated file; ask Sume support if your use depends on it.
| Question | Answer |
|---|---|
| Does Google say Lyria audio is watermarked? | Yes: SynthID, all generated audio |
| Does Sume's music doc mention it? | No |
Which engine does sume/music-auto use? | Lyria 3.5 today |
What else does Google limit?
Google's page says safety filters block requests for specific artist voices or copyrighted lyrics, and its model card points to Google's Generative AI Prohibited Use Policy. Sume's docs mention policy rejections on the music route: revise the flagged content while keeping the musical brief.
Sources
Related posts
More in Models
- AI sky replacement: swap the sky and keep the photo
AI sky replacement: send the photo to an image-edit model, describe only the new sky, and list what stays. Then check roof and tree edges. Cost and limits.
- Why AI video models do not lip sync to a voiceover, and what to use
Sume docs: video models do not lip-sync to generated speech or a voice-over. For a talking face use Avatar video or the lip-sync endpoint (still plus audio).
- Batch edit photos with AI: one edit across many photos
To batch edit photos with AI, send the same edit prompt once per photo, with that photo as the reference. How to script it, pace it, and what it costs.
- AI outfit change video: change clothes in a clip with AI
Change someone's outfit in an existing video with an AI video-to-video edit: name the garment, what it becomes, and what must stay the same.
Written by Sume