MELO音乐 — Chinese AI Song Generator with Vocals and Voice Cloning

What is MELO?

MELO (MELO音乐) is a Chinese AI music-generation platform operated by Beijing Lingxi Shenzhi Technology Co., Ltd. Its pitch is summarized by its tagline: "write songs even without knowing music theory." Users supply a text prompt, an image, or a mood description, and the platform returns a complete song with vocals, lyrics, and a professional arrangement. Creation is conversational — a song can be described in dialogue with zero learning curve — and the service is optimized first for Mandarin and Cantonese, with additional support for English, Japanese, Korean, and Spanish.

The platform runs in the browser, as iOS and Android apps, and as a WeChat mini program with cross-device sync, and it is clearly aimed at China's short-video and self-media ecosystem: finished tracks can be shared one-click to Douyin and Kuaishou with visual templates.

How MELO Works

Generation follows three steps. First, input inspiration — text, an image, or an emotion description. Second, the platform's "Melo engine" performs what the site describes as real-time melody computation. Third, the song is delivered as a lossless, high-quality track in one click.

Beyond one-shot generation, MELO supports several deeper workflows. Covers and remakes let users upload original audio or an acapella for an AI cover, backing-track extraction, or rearrangement. Voice cloning accepts a recording so songs can be sung in the user's own voice. Songwriting continuation takes a half-written chorus and returns AI continuation options. A member center manages creations, and a discovery feed surfaces other users' AI tracks.

Main Features

• Multi-input song generation — text prompts, images, or mood descriptions become complete songs with vocals, lyrics, and arrangements.
• Conversational creation — describe the song you want in dialogue; the site markets this as a zero-learning-curve path for users without music theory.
• Covers and remakes — upload audio or acapella for AI covers, backing-track extraction, or rearrangement.
• Voice cloning — record yourself once and have generated songs sung in your own voice.
• Songwriting continuation — submit a half-written chorus and receive AI continuation options.
• Multi-track lossless export — up to 12 lossless stems for post-production, plus one-click Douyin/Kuaishou sharing with visual templates.

Pricing

MELO does not publish prices. Neither the homepage nor its pricing/membership page lists plan prices, generation quotas, or free-tier limits — the page describes membership qualitatively as unlocking "more super creation permissions" and compute priority for long lossless output. What is documented is the rights model: a free tier exists, but free users do not automatically own their songs — they can purchase "creation certification" (创作存证) to gain ownership. Paid members' songs belong to the user, with free commercial use and distribution to music and short-video platforms. In short: creating is possible for free, but owning and monetizing requires payment, and the amounts are only visible inside the product.

Common Use Cases

Testimonials and positioning on the site point at several concrete audiences:

• Vloggers — background music without copyright headaches for video channels.
• Podcasters — custom theme music for shows.
• Independent musicians — breaking creative blocks with generated melodies and lyrics.
• Students and beginners — producing lyrics and melodies without theory background.
• Gift and event songs — personalized tracks for birthdays, anniversaries, and weddings.
• Voice-heavy creators — audiobook narrators using voice cloning, and DJs blending styles such as "cyberpunk + folk."

Limitations

The biggest gap is commercial transparency: no plan prices, quotas, or free-tier generation limits are published anywhere, so cost and volume cannot be planned before signing up. Rights are gated — free users do not own their output unless they buy certification, and commercial use requires membership. The product is China-first: sharing targets Douyin/Kuaishou, distribution runs through the WeChat mini program and APK for Android (not the Play Store), and the site itself is in Chinese, which raises the bar for non-Chinese speakers despite the multi-language song output. Finally, quality and speed claims come from the operator's own marketing and user testimonials; there are no published quotas, failure policies, or third-party benchmarks to verify them.

Summary

MELO is a feature-complete Chinese AI music studio for people who want finished songs rather than raw audio: multi-input generation, conversational creation, covers, voice cloning, continuation, and 12-stem lossless export, wrapped in a short-video-friendly sharing pipeline. Its weaknesses are equally concrete — opaque pricing, paywalled ownership, and a China-first distribution model. For Chinese-speaking creators making BGM, theme songs, or personalized gift tracks, it covers the whole path from idea to shareable song; buyers outside that ecosystem should weigh the language, platform, and rights constraints before committing.