Music

Full Songs with Vocals

Create complete tracks with AI-generated lyrics and vocals in multiple languages including English, Spanish, German, and Japanese. The model understands nuanced prompts and can generate songs in most styles or genres.

Instrumental Tracks

Generate purely instrumental music across any genre, from cinematic scores to ambient lo-fi beats. Perfect for background music, film scores, or any project requiring instrumental accompaniment.

Specific Song Structures

Use sectional descriptions in your prompt to build songs piece by piece, defining the Intro, Verse, Chorus, Breakdown, and Outro. This gives you granular control over your song’s composition and flow.

Music for Media

Design custom soundtracks for videos, advertisements, games, or other media by describing the scene or mood. For example: “A high-intensity orchestral track for an epic battle scene” or “Upbeat corporate jingle for a tech startup.”

Genre-Specific Music

Generate highly specific styles by including detailed prompts, such as “Traditional Spanish flamenco with palmas, nylon guitar, and Spanish-language vocals” or “1980s synthwave with analog synthesizers and retro drum machines.”

Adding and Removing Sections

Adding a New Section:

• To insert a section between existing ones, hover over the section in the section view and click the ”+ Add Section” icon that appears. This will add a section after the current section.
• To add a section at the end of your track, scroll to the end of the timeline and click the ”+” button.
• Drag the new section in the timeline to adjust its duration.

Removing a Section:

• Hover over the section you wish to remove in the song structure view.
• Click the delete icon (X) that appears in the corner of the section block.

Editing Lyrics and Prompts

To change the lyrics or instrumental prompts of any existing section:

• Click inside the text box for that section (e.g., Intro, Main Theme).
• Type your new lyrics or edit the existing prompt.
• Use bracketed descriptions like “[energetic guitar solo]” or “[drum fill]” for instrumental parts.

Style Control

For advanced control over specific musical elements:

• Hover over the section you want to edit and click “Edit styles of this section”.
• In the “Section styles” window, you can:
  • Include styles: Add specific musical characteristics like “gradual filter cutoff”, “hi-hats fade out”, or “long delay feedback on vocals.”
  • Exclude styles: Prevent certain elements like “abrupt ending” or “new elements.”
• Click Save to apply these style rules to that specific section.

Direct Prompting

Use the conversation interface at the bottom of the editor to make specific changes with natural language:

• Type direct instructions like “Make the chorus more energetic” or “Add a guitar solo after the second verse.”
• This allows for creative editing beyond the structured tools.

Regenerating Changes

After making any edits—whether adding, deleting, or modifying sections—your changes are staged but not yet applied to the audio.

• Your edits will not take effect until you click the Generate button.
• Once you click Generate, the model creates a new version of the song incorporating all your changes.
• Feel free to experiment with different combinations of lyrics, styles, and structures between generations.

Be Specific with Genre and Style

Instead of generic terms like “rock music,” try detailed descriptions:

• “Energetic 1980s synth-pop with a driving drum machine beat and male vocals”
• “Melancholic indie folk with fingerpicked acoustic guitar and ethereal female harmonies”
• “Heavy metal with palm-muted riffs, double bass drums, and aggressive vocals”

Layer Multiple Descriptors

Combine mood, instrumentation, tempo, and use case for better results:

• “A slow, melancholic piano melody over ambient synth textures, suitable for a tragic film scene”
• “Upbeat corporate jingle with bright synthesizers, punchy drums, and an optimistic melody”
• “Dark, atmospheric electronic track with deep bass, glitchy percussion, and haunting vocal samples”

Define Instrumentation

Call out specific instruments you want to hear:

• “Upbeat funk track with a prominent slap bass line, funky rhythm guitar, and a horn section”
• “Classical string quartet with violin, viola, and cello”
• “Jazz ensemble with piano, upright bass, brushed drums, and tenor saxophone”

Use Include/Exclude Styles

Refine your output by explicitly including or excluding certain elements at the track or section level:

• Include: “acoustic,” “four-on-the-floor kick”, “reverb-heavy vocals”, “analog warmth”
• Exclude: “repetitive structure”, “electronic elements”, “abrupt ending”, “distorted vocals”

Build Section by Section

For maximum control, start with a short generation (e.g., 30 seconds for an Intro) and build your song piece by piece.

1. Generate the Intro and refine until satisfied.
2. Click ”+ Add Section” to add the next part.
3. Specify the style and content for the new section (e.g., Verse, Chorus, Bridge).
4. Use the conversation interface for specific instructions on each part.
5. Build your track progressively, ensuring each section flows naturally.

Iterate and Refine

Don’t start over if the first generation isn’t perfect. Small changes can have a big impact.

• Adjust your prompt and regenerate specific sections.
• Use the editing tools to fine-tune individual parts.
• Experiment with different style combinations.

Export Formats

Generated audio is provided in MP3 format with professional-grade quality (44.1kHz, 128-192kbps). Other audio formats will be supported soon.

Quality Settings

Export quality varies by subscription tier:

• Free, Starter, and Creator: Standard quality exports.
• Pro, Scale, Business, and Enterprise: High-fidelity, studio-grade exports. All exports maintain the full dynamic range and frequency response of the original generation.

Sharing

Use the Share button to:

• Generate a shareable link to your song.
• Customize the visualizer that accompanies your track.
• Share your creations with collaborators or audiences.

The cost for generating music is calculated upfront based on your selected settings (like duration) and is shown next to the Generate button. Unlike other ElevenLabs tools with fixed credit pricing, the number of credits used can vary depending on your subscription plan. This is because the cost is based on a fixed fiat price, which translates to a different number of credits for each plan.

When using the Auto duration setting, the cost is fixed for that generation attempt, regardless of the final length of the song. This means a 2-minute song generated on “Auto” costs the same as a 3-minute one, providing predictable upfront pricing.

You can create everything from short clips to full-length tracks. For maximum control, you can start with a short duration like 30s to create an initial section, then iteratively use the ”+ Add Section” button to build out your song piece by piece, extending it to your desired length.

The minimum duration of a song is 10 seconds and the maximum is 5 minutes.

Eleven Music is cleared for broad commercial use. See our Music Terms for specific usage details per subscription tier.

Eleven Music supports vocals in multiple languages including English, Spanish, German, and Japanese. The model can generate lyrics in these languages, or you can provide your own lyrics for the AI to sing.

You can edit individual sections, modify lyrics, adjust durations, add or remove sections, and fine-tune the style of specific parts. Changes take effect when you click Generate to create a new version.

The key is detailed, specific prompting. Include genre, mood, instrumentation, tempo, and use case in your descriptions. Use the Include/Exclude styles feature for fine control, and build songs section by section for maximum precision. See our Prompting Guide for comprehensive tips.