Alongside images and video, Imaginations supports audio generation — including music, atmospheric sound, and AI voice clips. This is separate from Voiceovers (which uses your own custom voice library for narration); audio in Imaginations is generative, prompt-driven content.
Opening the Generate dialog for audio
- Click Imagine on the Imaginations page.
- In the Generate dialog, select Audio as the output type.
What you can generate
- Music — original compositions in any genre, mood, or instrumentation
- Atmospheric sound — ambient environments, sound effects, backgrounds
- Voice clips — generated AI voices for specific lines or narration
Writing an audio prompt
For music, describe:
- Genre (e.g. "lo-fi hip hop", "orchestral score", "upbeat electronic")
- Mood (e.g. "melancholic", "energetic", "calm")
- Instrumentation (e.g. "piano and strings", "synthwave bass and drums")
- Tempo or pace
For voice, describe:
- Character and tone (e.g. "warm, friendly female narrator", "deep authoritative male voice")
- Delivery speed and style
- The text you want spoken (include it directly in the prompt)
Generating lyrics (for music)
Before generating, click Generate lyrics in the dialog to use Euryka's lyrics helper. Describe the theme, story, or mood and Euryka drafts lyrics you can edit and include in your music prompt.
Enhancing your prompt
Click Enhance prompt to let Euryka refine your audio description automatically. This is particularly useful for music where precise terminology (BPM, key, chord progressions) can significantly improve results.
Reviewing and downloading
Generated audio appears as iterations in the current session. Click an iteration to preview the audio in the lightbox. From there you can:
- Play the clip
- Download the audio file
- Remix — generate a variant with tweaked parameters