Create tracks, add vocals, mash up and cover — all via AI. Go to generation →
Suno V5.5 — новейшая модель. Максимальное качество вокала, инструментов и аранжировки.
Each generation creates two tracks — pick the best or try both.
3 models (ChatGPT, Gemini Pro, Grok) refine lyrics, style and tags to a professional level.
Generation error? Credits are auto-refunded. You only pay for results.
Create from scratch, make covers, extend, or work with existing audio files.
Напишите текст песни, задайте стиль через теги — Suno V5.5 создаст полноценный трек с вокалом и аранжировкой.
Upload a beat or instrumental — the AI overlays vocals with the chosen lyrics, style and voice gender.
Upload vocals or an a-cappella — the AI selects and adds an instrumental arrangement in the desired style.
Upload 2 audio files — the AI merges them into one track. Remixes, covers, sonic experiments.
Upload any track (up to 8 min) — the AI creates a cover version in a new style while keeping the melody. Change genre, voice, original weight.
Upload a track (up to 8 min) — the AI extends it while keeping style and mood. Specify the extension point or leave "from the end".
Extend tracks, replace sections and create voice profiles.
Continue a track from any point. New lyrics, new style — or simply "one more verse". Build long tracks step by step.
Don't like the chorus? Set a range (from 6s), write new lyrics — the AI rebuilds only that fragment.
Extract a voice from a track and reuse it. One voice for all songs — album, series, character.
Shape the sound with sliders and negative tags.
0 — loose interpretation, 1 — strict adherence to the given tags.
0 — classic, 1 — experimental choices and surprises.
0 — clear lyrics, 1 — melody and arrangement first.
Exclude genres, instruments, and effects: autotune, electronic drums, fade out
Built-in chatbot for every stage of music creation.
Practical tips for great tracks on the first try.
Tag your sections — Suno understands verse vs. chorus better that way. Sections should be short (4-8 lines). Long blocks = blurred vocals.
Available tags: [Verse], [Chorus], [Pre-Chorus], [Bridge], [Outro], [Intro], [Hook], [Break], [Interlude], [Drop], [Rap], [Whisper], [Scream], [Ad-lib], [Guitar Solo]
Start the prompt with a vocal persona description — this sets the tone from the first seconds. Performance hints can be added right inside section tags.
[Verse - falsetto, ethereal], [Bridge - spoken word], [Verse - guttural growl, aggressive]
Punctuation and special marks affect like how the AI sings the text:
In the "Style" field, list genre, mood, instruments — comma-separated — and BPM is required. Without BPM, the rhythm may not match the lyrics.
💡 Tip: Instead of artist names, describe the physics of the sound: "down-tuned palm-muted rhythm guitar, double kick drums" is more precise than "like Metallica".
List elements that should NOT appear in the track. This removes unwanted artifacts.
Tip: start without negative tags, listen to the result, and add specific exclusions if needed.
Professional approach to crafting the perfect prompt — don't try to write everything at once:
Lyrics with section markup or an idea description. The AI assistant will help.
Genre, mood, and instrument tags. Or press 'Enhance' — AI will pick them.
Suno V5.5 создаст 2 варианта. Выберите лучший.
Extend, replace a section, create a persona — and repeat.
| Action | Credits |
|---|---|
| Track generation (2 variants) | — |
| Extend track | — |
| Replace section | — |
| Cover (Upload & Cover) | — |
| Extend file (Upload & Extend) | — |
| AI prompt enhancement | from 1 cr |
| AI prompt editing | from 1 cr |
Suno V5.5 — новейшая модель генерации музыки с максимальным качеством звука, вокала и инструментов. По сравнению с V4 значительно улучшены чёткость вокала, разнообразие аранжировок и передача эмоций. Промпт до 5000 символов, стиль до 1000 символов.
A single generation produces 2 track variants — you can pick the best one. AI lyrics enhancement costs from 1 credit. If generation fails — credits are refunded automatically.
Yes. Enable "Instrumental" mode — the track will have no vocals. You can also do the reverse: upload a ready instrumental and add AI vocals via "Add Vocals" mode.
"Add Vocals", "Add Instruments", "Mashup", "Cover", and "Extend File" modes have an audio upload field. MP3, WAV, FLAC, and OGG are supported (up to 50 MB, up to 8 minutes).
A mashup combines two audio files into one track. Upload two audio files, set a description and style — the AI will create a unique blend. Great for remixes and experiments.
Cover (Upload & Cover) — upload any track, set a new style, and the AI will create a cover version while keeping the original melody. You can change genre, voice, and the original's influence (Audio Weight).
Extend File (Upload & Extend) — upload your track and the AI will seamlessly continue it. Set the extension point (seconds) or leave 0 to continue from the end.
Open the finished track and click "Extend". Specify the position (in seconds) to continue from and, optionally, new lyrics. This lets you iteratively build long compositions.
Open the finished track → "Replace Section". Specify the start and end of the segment (minimum 6 seconds), write new lyrics — the AI will redo only that segment, leaving the rest untouched.
A voice persona is a voice profile extracted from a generated track. You can reuse it across new tracks for a consistent sound — handy for series, albums, and characters.
Three sliders: "Style Weight" (how strictly to follow the style), "Weirdness" (how experimental the sound is), and "Audio Focus" (balance between lyric clarity and melody). There are also negative tags to exclude unwanted elements.
Use tags like "death metal, guttural vocal, false cord growls". Add phonetic hints in the lyrics (grrr, aaargh). Set the "Weirdness" slider to 0.6–0.75. For best results — generate the instrumental separately, then add vocals via "Add Vocals".
Add "(Do not change any words)" before the lyrics section. Also helpful: duplicate the hook/chorus exactly in each [Chorus], shorten sections to 4–6 lines, specify BPM in the style for a clear rhythm.
Tracks are stored for 14 days. We recommend downloading ready tracks immediately — every card has a download button.
ChatGPT, Gemini Pro, Grok и бесплатная FreeGPT.
Yes, always. If generation fails for any reason, credits are automatically returned to your balance.
We use cookies to operate the service, keep your session, and collect anonymous statistics. See our Privacy Policy.