A deep dive into every generation setting and what it controls.
Every setting you provide is sent to the AI as part of the musical brief. The more specific you are, the more consistent and on-target the result.
The musical key (e.g. G, Bb, F#) and scale (Major or Minor) set the tonal foundation. A Minor key has a darker, more emotional feel; a Major key feels brighter and more resolved.
Tip: if you recorded something in Larka, your note is already pre-filled with the detected key and scale.
Genre tells the AI what style to compose in. Larka includes 25 genre presets you can pick from:
Pop, Rock, Jazz, Blues, R&B, Folk, Electronic, Classical, Reggae, Country, Latin, K-Pop, Afrobeats, Bossa Nova, French Pop, Bollywood, House, Techno, Ambient, Lo-Fi, Punk, Indie, Funk, Soul, Soundtrack.
Genre starts unset on new notes — pick one before generating for best results. You can also leave it empty and let the AI choose based on your lyrics and mood.
Mood adds emotional color on top of the genre. Examples: melancholic, uplifting, dark, dreamy, energetic, romantic, nostalgic, tense.
Combining genre and mood gives the most directed results: "Folk · melancholic" vs "Folk · uplifting" produce noticeably different songs.
Set a target tempo in beats per minute. Leave blank to let the AI choose. Common ranges: 60–80 for slow ballads, 80–110 for mid-tempo, 120–140 for uptempo, 140+ for fast/dance.
If you want lyrics in a specific language, set it here. Set to "Auto" to let the AI decide (usually English). Works with most languages.
List specific instruments you want featured, separated by commas. Examples: "acoustic guitar, piano, light drums", "strings, cello, violin", "electric guitar, bass, drums".
Leave blank to let the AI choose instruments that suit the genre.
Describe the vocal character: "Female · Breathy", "Male · Deep", "Choir", "Rap", "Spoken word". Only applies when Instrumental is off.
Two sliders in Advanced, available for both notes and recordings. Both start on Auto, which lets the AI decide. Move a slider to set it yourself, or tap "Auto" to reset it.
Style influence: how strictly the song follows your genre, mood and instruments. Lower gives the AI more freedom; higher sticks closely to what you asked for.
Weirdness: how experimental the result gets. Lower stays safe and familiar; higher takes more creative risks.
When you generate a song from a recording or imported audio, Audio influence controls how closely the new song follows your audio. Higher keeps more of your melody and timing; lower treats your recording as loose inspiration. It doesn't appear on notes, since notes have no audio.