Cycling through all six. Tap any point to stop.

Measured on six things

AI music generator that turns text into editable full songs.

Security

HTTP probe scored 2/13: HTTPS not enforced, no HSTS, CSP, frame protection or secure cookies. Only SPF and DMARC present; no SOC 2 or bug bounty found.

Screenshots

1 / 1

Doremi

About this app

Doremi is an AI music workspace for creating songs from prompts or lyrics. Choose from supported music models, set vocal or instrumental output, and review example tracks. Use AI Lyrics for songwriting and Doremi Agent to explore ideas; project templates support batch creation and comparing models or styles. After generation, tools let creators extend and remix songs, make mashups, replace sections, recognize lyrics, and split stems. Doremi offers a free plan alongside paid plans.

Key features

1Multi-model generation: run Suno, Mureka and ElevenLabs side by side on the same prompt and compare results.
2Full editing suite: extend, remix, cover, replace sections, mashup and AI mastering on generated tracks.
3AI stem splitter and vocal remover for isolating vocals, drums, bass and instruments from any song.

Target customer profile

Drawn from the product itself, not from a survey.

Scout Review

Written by AI from measured evidence, scored out of 100.

Doremi is a competent, well-designed AI music workspace with a genuinely useful multi-model pitch and a full editing suite behind it. The design is polished and the accessibility audit is strong at 95/100. But the fundamentals have holes: a 15.3-second mobile LCP, a 2/13 security-header score with HTTPS not enforced, and a free tier so thin it barely functions as a trial. The idea is sound and the category is hot; execution on the boring infrastructure is where it needs work.

Measure by measure

Usability

Five pricing tiers with clear credit/song allowances and a stated 3-step flow. FAQ covers licensing and editing, but answers are collapsed and onboarding depth is unverified.

Design

Polished dark-themed music workspace with community track cards, genre filters and clear model badges. Dense but coherent landing page; branding is consistent across homepage and pricing.

Speed

Lighthouse mobile 66/100 with a brutal 15.3s LCP; desktop 72/100. CLS 0 and TBT 217ms are fine, so the hero media is the anchor dragging the page down.

Security

HTTP probe scored 2/13: HTTPS not enforced, no HSTS, CSP, frame protection or secure cookies. Only SPF and DMARC present; no SOC 2 or bug bounty found.

Accessibility

Lighthouse accessibility 95/100 with one failing audit: viewport disables user scaling. No WCAG statement found; language selector hints at multi-language support.

Growth

Real, growing category with a genuine wedge: one workspace aggregating Suno, Mureka and ElevenLabs plus a full editing suite. Early stage (domain ~0.6 years) but the idea has room.

Improvement areas

Security headers: HTTPS not enforced, no HSTS/CSP/frame protection, 2/13 on the probe checklist.
Mobile performance: 15.3s LCP drags the Lighthouse mobile score to 66/100.
Free tier is too thin (10 credits / 4 songs per month) to function as a real trial of the editing suite.

Design & Usability

Doremi's design is genuinely above-average for an indie AI tool: a dark, cohesive workspace with community track cards, genre filters and model badges that make the multi-engine pitch legible at a glance. The problem is density — the homepage stacks a hero, a community feed, an eight-tool grid, a model comparison, a three-step how-to, an FAQ and a final CTA, and the reader has to work to find the actual product. Usability is competent: five pricing tiers with explicit credit and song allowances, a stated three-step flow, and an FAQ that covers licensing and editing. But the FAQ answers are collapsed, and the free tier's 10 credits/month (4 songs) is thin enough that 'start for free' is more of a sample than a trial. Onboarding depth beyond the marketing copy is unverified.

Speed & Security

Speed is the weak spot. Lighthouse mobile scores 66/100 with a 15.3-second LCP — that is a genuinely bad first impression for a page whose whole job is to get someone to click 'Start for Free.' Desktop is 72/100, and CLS 0 with TBT 217ms tells you the layout is stable and the main thread is fine; the bottleneck is heavy hero media. Security is worse. The HTTP probe scored 2/13: HTTPS is not enforced, and there is no HSTS, CSP, frame protection, X-Content-Type-Options, Referrer-Policy, Permissions-Policy or secure-cookie flag. Only SPF and DMARC are present. No SOC 2, ISO, pen-test disclosure or bug bounty was found in the site content, and privacy-policy depth could not be verified. For a product that handles user-generated audio and payment tiers up to $129.99/month, that header posture is hard to defend.

Accessibility & Growth

Accessibility is the strongest measured dimension: Lighthouse scores 95/100, with a single failing audit — the viewport meta disables user scaling, which blocks pinch-zoom for low-vision users. That is a one-line fix and a real flaw, but the rest of the automated checks pass cleanly. No WCAG conformance statement was found, and a language selector hints at multi-language support without confirming it. Growth is a judgement on the idea: AI music generation is a fast-growing category, and Doremi's wedge — one workspace that runs Suno, Mureka and ElevenLabs side by side, then lets you extend, remix, cover, replace, split stems and master — is a real differentiator against single-model competitors. The category has headroom and the editing layer is under-owned. The limit is that the underlying models are rented, so the moat is workflow, not technology, and a well-funded incumbent could replicate it. Domain age (~0.6 years) is context only.

Conclusion

If you are shopping for an AI music tool and you care about comparing models and editing the output rather than just generating a clip, Doremi's workflow is worth a look. Go in with eyes open: the free tier is a sample, the site is slow to load, and the security posture is not yet something you would want to put a client's unreleased track behind. Fix the headers, fix the LCP, and this becomes a much easier recommendation.

Suggestions for improvement

Enforce HTTPS and add HSTS, CSP, frame protection and secure-cookie flags — the 2/13 header score is the biggest fixable gap.
Cut the 15.3s mobile LCP by compressing or lazy-loading the hero media; target sub-3s.
Remove user-scalable=no from the viewport meta — it is a one-line accessibility fix that clears the only failing audit.
Expand the free tier or add a real time-limited trial so 'Start for Free' demonstrates the editing suite, not just 4 songs.

How we score

Alternatives comparison

Named competitors, point by point. Nobody paid to appear here or to be left out.

	Doremi
Multi-model AI music generation	Yes — Suno, Mureka and ElevenLabs side by side	Same product (self-listed alternative)	Not described
Post-generation editing (extend, remix, cover, replace)	Yes — full music editor with extend, remix, cover, replace, mashup	Same product (self-listed alternative)	Not described
Stem splitting and vocal removal	Yes — AI stem splitter and vocal remover	Same product (self-listed alternative)	Not described
Free plan	Yes — 10 credits / 4 songs per month	Same product (self-listed alternative)	Not described
Commercial license included	Yes — commercial license and downloadable certificate on all paid tiers	Same product (self-listed alternative)	Not described