Sonauto Now Available on fal

Team fal

• 2 min read
Sonauto Now Available on fal

Today, we're excited to announce our partnership with Sonauto, bringing their latest v2.2 music generation model to the fal platform on day one. This integration delivers the highest quality vocals and more creative instrumentation with more depth than other music models—all accessible through fal's simple, developer-friendly APIs.

What is Sonauto v2.2?

Sonauto v2.2 sets a new standard for AI music generation with CD-quality 1.5-minute tracks featuring superior vocal clarity and more creative instrumentation with more depth. Listen to the remarkable difference between an airy K-pop vocalist and a belting rock ballad—the model captures the nuanced characteristics that define each genre.

Check out the endpoints here:

Key features include:

  • Superior Vocal Quality: Industry-leading vocal synthesis that distinguishes between different singing styles
  • More Creative Instrumentation: Richer, more layered arrangements with greater depth than other models
  • BPM Configuration: Manually set tempo for precise control (new in v2.2)
  • Song Extension: Extend 1.5-minute clips into full-length songs
  • Audio Editing: Upload and modify existing tracks—extend or replace specific sections
  • Text Prompt Mode: Automatic generation of styles and lyrics from simple descriptions

Try it out on fal

Sonauto v2.2: Generate 1.5-minute CD-quality music from lyrics and style descriptions. Features the best vocal quality available, more creative instrumentation with more depth, and manual BPM configuration (new in v2.2).

Hear It in Action

Example #1: Let us Cook
Prompt:  tropical house, 2020s, edm, dance pop, melodic house, pop

0:00
/2:00

Example #2: Cow
Prompt: electro house, electronic dance, classical crossover, future

0:00
/2:35

Example #3: Cooked Alright
Prompt: philly soul, funk, rhythm and blues, urban, soul, 1970s

0:00
/1:35

Example #4: Rocks
Pcountry pop, country rock, contemporary country, heartland rock

0:00
/1:35

Example #5: IDK
Prompt: electropop, pop, contemporary r&b, latin pop, dance pop, 2020s

0:00
/1:35

Getting Started with High-Quality Results

Creating Optimal Lyrics Structure

The model generates 1.5-minute clips. For best results, provide approximately four verses of lyrics:

  • Chorus
  • Verse 2
  • Pre-Chorus
  • Chorus
  • (Optional) Post-Chorus

Each verse should contain around four lines to properly fill the generation window.

Using Style Tags vs. Text Prompts

Direct Input Mode:

  • Select specific style tags for genre, mood, and instrumentation
  • Provide your own lyrics
  • Optionally set BPM manually (new in v2.2)

Text Prompt Mode:

  • Submit a natural language description
  • The model generates appropriate style tags
  • Optionally generates lyrics following best practices

Extending and Editing Tracks

  • Generate initial 1.5-minute clips
  • Use the extension feature to create full songs
  • Upload existing audio to replace or extend specific sections

What to Know About Model Specifications

Output Format

  • Duration: Always 1.5 minutes per generation
  • Quality: CD-quality audio (44.1kHz, 16-bit)
  • Cost: 7.5 cents per generation

Language Support
Currently known to support lyrics in English, Spanish, French, and German. Other languages may work but are untested.

Start Creating Today

All of this is ready to use on fal —just plug and play via API. Get started with Sonauto v2.2 today and bring professional music generation to your creative workflows.

Explore the model in the fal model gallery.