Music, voice-over and audio

What Eddie can do with audio

Eddie can add:

• AI-generated music — describe a style or mood and Eddie generates a new track for your edit. On the timeline, the empty Music lane also has a one-click ✨ Generate button with style, mood and tempo pickers.
• Your own music — on the Music lane, click + and choose Upload a file (MP3, WAV or M4A, up to 150 MB) to place it at the playhead, or upload it to your workspace's music shelf and pick it from there.
• Several music cues — the Music lane can hold more than one track, one after another, with an automatic 1-second crossfade where they meet.
• Sound effects — generated whooshes, hits and pops on the SFX lane. Motion graphics place their own cues, and you can add one at the playhead with the lane's ✨.
• AI-generated voice-over — describe the script and tone you want, or let Eddie draft the script from the edit. You can preview the voices before generating.
• A dubbed voice track — translate the edit's dialogue into another language and speak it on the voice-over lane. See Dub your edit.
• Automatic ducking so music doesn't fight with dialogue — on by default, and switchable per cue.
• Cuts on the beat — ask Eddie to "snap the cuts to the beat" and it measures the music's beat grid, then nudges cut points onto beats without clipping a spoken word. Measuring an uploaded track costs 2 credits per minute of audio; generated tracks already carry their grid.

For step-by-step help with all of these, see Generate with Eddie.

Mute a track

Every audio-bearing lane on the timeline has a speaker toggle on its left rail:

• V1 (your clips) — mutes the clips' own recorded audio. This is how you build a voice-over-narrated sequence over A-roll: the picture plays, the camera sound doesn't.
• VO — mutes every voice-over clip.
• SFX — mutes every sound effect.
• Music — mutes the soundtrack.

A mute is part of the edit (everyone on the project sees it, and it survives reload), and it carries into your exports:

• MP4 renders leave the muted track out of the mix — muting the clips' audio renders their sound as silence, with the picture and timing untouched.
• Premiere project (.prproj) arrives with the audio tracks fader-muted (the mixer's M button) — the media is all there, un-mute to bring it back. Premiere XML ships those audio clipitems disabled.
• Formats with no track mute Eddie can set (FCPXML, OTIO, DaVinci .drp) include the audio and say so in the export notes — mute the track in your NLE to match.

Muting is non-destructive: nothing is deleted, and toggling the speaker back restores the sound everywhere.

Even out the dialogue: Auto-level speech

Auto-level speech (−16 LUFS) brings each A-roll source's dialogue to the same level before the mix. A cut between a quiet speaker and a loud one then sounds even.

It is a checkbox on the Video file, Premiere Pro, DaVinci Resolve and Final Cut Pro destinations of the Export menu. When you have not chosen, the default depends on where the audio goes:

• Video file (MP4) — on when the edit uses more than one A-roll source, such as multicam angles or several interview files. Off for a one-source edit.
• Premiere Pro, Final Cut Pro and DaVinci Resolve timelines — off. The clip volumes in your timeline stay as recorded unless you tick the box.

Your choice is saved with the edit, and one choice covers every destination. If you tick it for an MP4, your next NLE export is levelled too.

Set the final loudness of an MP4

The Video file destination also has a Final loudness choice. It sets the level of the whole finished mix — dialogue, music, voice-over and effects together:

• Web & social (−14 LUFS) — the default. YouTube, Spotify, Instagram and TikTok play back near this level.
• Broadcast EBU R128 (−23 LUFS) — Europe and most of the world.
• Broadcast US ATSC A/85 (−24 LKFS) — US television.
• Off (keep the mix level).

Final loudness applies to MP4 renders in the cloud. It does not change a timeline export.

Ask Eddie to set them

You can also set both in the chat: "level the dialogue", "turn off auto-level", "make the final loudness broadcast EBU". Eddie saves the same setting the Export menu shows. A setting applies to the next render or export, so export again to get a levelled file.

What Eddie can't do

• Your own sound-effect files — sound effects are generated; you can't upload an SFX file to the lane.

If you need that in the final piece, export the edit (MP4, or a timeline via FCPXML/Premiere XML) and add them in your usual editor.

Cost

Generating AI music, sound effects or voice-over is a paid operation billed in credits (voice-over on the script length, music and sound effects on their length). Placing an uploaded track from your music shelf is free, and so is reusing a sound effect your workspace already has. See what it costs for the numbers. If you're unsure, ask Eddie "how much would this cost?" first — it quotes before it spends.