Lyria 3 Music Generation in Gemini App: 6 Prompting Tips
Key Points
- Generate 30-second original music tracks from text, images, or videos
- Support for custom lyrics, genre blending, and vocal characteristics
- Export tracks as MP3/MP4 for easy sharing across platforms
Summary
Lyria 3 is Google DeepMind's latest generative music model integrated into the Gemini app, enabling users to create original 30-second tracks from text, images, or videos. The model supports various genres, instruments, vocals, and custom lyrics.
Key Points
- Text-based prompts: Create tracks with specific themes, genres, and vocal styles using descriptive text prompts
- Visual input support: Upload images or videos as inspiration for musical generation
- Genre flexibility: Supports multiple genres and eras, including genre blending (e.g., K-pop with Motown)
- Instrument customization: Specify instruments or let Lyria auto-select based on genre
- Vocal control: Define vocalist characteristics (male/female, baritone/soprano, vocal qualities)
- Custom lyrics: Use "Lyrics:" prefix for custom text or let Lyria generate thematic lyrics
- Export options: Download as MP3/MP4 for sharing on social media or messaging