mayk.ai

Model Configuration

Model Version

Genre Description

Text containing genre tags that describe the musical style (e.g. instrumental, genre, mood, vocal timbre, vocal gender)

Lyrics

Lyrics for music generation. Must be structured in segments with [verse], [chorus], [bridge], or [outro] tags

Number of Segments

Number of segments to generate (1-10)

Max New Tokens

Maximum number of new tokens to generate (500-40,000)

Temperature: 0.75

Controls randomness: Lower values produce more predictable output, higher values more creative

0.1 (More predictable)0.8 (Balanced)1.5 (More random)

Vocal Prompt URL

URL for a vocal prompt to guide the music generation (when provided, beat prompt is also required)

Drop your vocal file here or click to upload

Beat Prompt URL

URL for a beat prompt to guide the music generation (when provided, vocal prompt is also required)

Drop your beat file here or click to upload

Continue Reference

When enabled, the model will try to continue from the reference audio

Teacher Forcing

Teacher forcing type to use during generation