Voice-over music bed
Prompt: Keep speech in front while the music breathes between lines.
A playable ducking demo source pair.
automatic audio ducking
Automatically duck music under voice
Mix an existing voice track and music bed through the Sonilo audio-ducking API for automated speech-first mixing.
Duration
Match foreground and background assets
Prompt
Not required for audio ducking
Model version
Sonilo audio ducking
Input format
Voice audio/video plus music audio
Audio-ducking API
API workflow for mixing existing voice and music tracks. An API key and eligible API access are required.
POST /v1/audio-ducking accepts a voice track and a music track. Submit files or public URLs through the API, then poll GET /v1/tasks/{task_id} for the mixed output.
Need a music track first?
Generate music from textGenerate music from videoDurationMatch voice/music duration
Model versionSonilo audio ducking
Examples
Playable Sonilo results
Narrated product demo mix
Prompt: Lower the bed under narration, then restore energy during the reveal.
A commercial clip context for speech-first music balancing.
Tutorial background balance
Prompt: Keep the background audible but out of the way during explanation.
A video context for automatic ducking around spoken guidance.
Compare
Original, silent, and Sonilo output
Original video
The source clip keeps its original camera audio for reference.
Silent version
The same clip muted, so timing and visual beats are easy to judge.
Sonilo output version
A finished video mix where music is balanced against speech and picture.
Use cases
Built for this workflow
Narrated product demos where music should sit under speech automatically.
Podcast clips, trailers, and social edits with voice-over and background beds.
Tutorials, explainers, and course videos where dialogue clarity is the priority.
Batch post-production pipelines that need consistent speech-first mixes.
Developer workflows calling POST /v1/audio-ducking for async mix output.
FAQ
automatic audio ducking
What is automatic audio ducking?+
Automatic audio ducking lowers background music when speech is present, then raises it between phrases so narration stays clear.
Does audio ducking generate new music?+
No. Audio ducking mixes existing voice and music assets. Use text-to-music or video-to-music to generate a new soundtrack.
Can I duck music under a video voice track?+
The Sonilo API supports foreground speech or dialogue inputs and background music inputs for an async ducking workflow.
Where do developers call automatic ducking?+
Use POST /v1/audio-ducking and poll the returned task until the mixed output is ready.
Keep speech clear without hand-keyframing volume.
Use Sonilo audio ducking when you already have voice and music and need an automatic mix.