Models
Audio Models
Every music, voice, and audio model available on Kolbo.AI
Kolbo.AI covers the full audio stack: full songs, instrumental scores, licensed library tracks, expressive text-to-speech, and voice cloning and design.
Music
| Model | Provider | Type | Key spec | Best for |
|---|---|---|---|---|
| Suno | Suno | Music generation, vocal and instrumental | 52 style presets, custom mode with lyrics, full-length tracks | Complete songs with vocals or instrumental |
| Google Lyria 2 | Non-lyrical music | Up to 30s, instrumental only | Short-form soundtracks, intros and outros, social clips | |
| Minimax | Minimax | Music generation, vocals and instrumental | Tight loops, MP3 output | Electronic, ad cuts, short social music |
| Sonauto | Sonauto | Music generation | v3 produces 2 to 4 minute tracks, v2 around 1:35 | Longer-form music tracks |
| Synci | Synci | Licensed music library | Fully cleared catalog, browse by mood, genre, and BPM | Royalty-free professional tracks with no copyright risk |
| Vision Music | Suno, Lyria, Minimax (auto-routed) | Image or video to music | Auto-selects the engine from the visual mood | A soundtrack that matches a frame or clip |
Voice and Speech
| Model | Provider | Type | Key spec | Best for |
|---|---|---|---|---|
| Gemini Text-to-Speech | Text to speech | 30 voices, 40+ accents and languages, Voice Direction | Expressive multilingual narration with tone control | |
| Custom Voice (Clone) | ElevenLabs | Voice cloning | Clone from a short sample, reuse across TTS | Reusing a real or reference voice |
| Custom Voice (Design) | ElevenLabs | Voice design from a text description | Describe age, tone, and accent to invent a voice | Creating a custom voice with no sample |
Notes
- Suno is the flagship for full songs with vocals, with 52 style presets and a custom mode where you supply your own lyrics.
- Synci is a licensed library, not a generative model. Every track is cleared for commercial use with no copyright risk.
- Vision Music turns an image or video into a matching soundtrack and picks the best music engine automatically.
- Gemini Text-to-Speech offers Voice Direction to steer tone and delivery across 30 voices and more than 40 accents.
- Custom Voice supports both cloning an existing voice and designing a brand-new one from a text description, both powered by ElevenLabs.
Where to use these
These models power the Music tool, Text-to-Speech, Voice Cloning, Text-to-Sound, and Transcription and Subtitles. See Audio Tools for the workflows.