Music with structure and lyrics, voiceovers in natural voices, sound effects, and foley generated from the motion in your footage. Every render drops straight into the mix.
Music, voice, effects and foley in one tab
Switch the AI panel to Audio and pick a card. Music comes with song structure, scripts are read in natural voices, effects come from a description, and foley is generated from the motion in your footage.
Structured songs, with lyrics in verse and chorus tags, up to three minutes
Clip-matched foley: impacts, footsteps and ambience scored to the frame
Every render lands in your library and drags straight onto an audio lane
From script to voiceover, segment by segment
The script editor takes your narration in timed segments. Pick a voice and generate the read in one pass: each line lands on the timeline where its segment starts.
Segments carry their timings, so the voiceover lays itself out on the lane
Re-generate a single line without touching the rest of the read
Prefer your own voice? The in-browser recorder drops takes straight onto a lane
Questions
Can it write music with lyrics?
Yes. Music models take a style prompt and an optional lyrics field with verse and chorus tags, so you can shape the song's structure.
What voices are available for voiceovers?
Natural text-to-speech voices you pick per generation. The model reads your script as written, so spell out numbers and acronyms the way you want them spoken.
What is clip-matched foley?
A model reads the motion in a video clip and generates sound to match it: footsteps, ambience, impacts. It saves hunting a library for the right effect.
Can I record my own voiceover instead?
Yes. An in-browser recorder drops your narration straight onto an audio lane, and generated and recorded sound mix together freely.