Last updated: 12 August 2026
Upload: Drop in your Guide and your Master voice-over. Most common formats supported.
Align: Select Align Audio. Narrly works out which takes in your Master VO match your Guide.
Export: Download a Multi-Track WAV with all takes, or a Fast-Track version chosen by Narrly.
You've got Guide voice-over on your timeline and a Master session recording to cut in; containing a variety of takes and reads. Aligning these takes means scrubbing through the Master, finding the corresponding reads that match your timeline, dragging them into place, and then repeating the process for all the VO in the script. Narrly does that tedious part for you.
The Guide should be an export of the reference voice-over track from your timeline. The Master VO is the raw recording, which includes every take. Narrly works out which takes in the Master belong to which segments of your Guide, and aligns them.
Audio or video. WAV, MP3, AIFF, M4A, MOV and MP4 all work. If you drop in a video, Narrly takes the audio out of it for you. You need two files per session: one Guide and one Master.
You can upload stereo, but Narrly is built for voice and so works best in mono. Stereo files are summed down, and every channel in the Multi-Track or Fast-Track WAV comes back mono. Narrly only works on a copy of your files, not the original sources, so it's entirely non-destructive.
It depends on how long your recordings are, so longer files take longer. Narrly shows you progress while it processes, so you can leave the tab open and get on with something else. The results appear as soon as it's finished running, providing a timeline representation of your audio.
Guide voice-over sections that Narrly can't match a Master VO take to, don't make it into the final exports; but you'll see everything it did find on the results screen, before you download. Any undetected VO is flagged in red on Channel 1, as part of the grey Guide voice-over track. Hovering over the blue segments in the results page, also shows a short snippet of each take.
Discrete 48 kHz, 24-bit WAVs. Your Guide sits on Channel 1, and the takes stack from Channel 2 downwards in time-code order, each positioned at the point in the Guide where it matches. There's also a Fast-Track option, where Narrly chooses two takes per line and creates a triple channel output. Fast-Track is an experimental feature and the results can vary considerably.
It's a standard multi-channel WAV, so import it the way you'd add any other file into your NLE or DAW of choice. Once it's in: mute, solo, cut and disable channels as normal in your timeline.
It's a head start, not a finished edit. Narrly makes mistakes and won't always place every take correctly, especially if the Master VO varies considerably from the Guide. It's meant to speed up a boring task, not to make the final call, so human oversight is always strongly advised.
It's processed on our own systems within the EU, and it's not shared with third-party AI platforms or cloud services. Your files are deleted when you close the tab or start a new session, and anything left over is purged within a few hours. We don't keep your audio, can't access your session or store the outputted files. Our Privacy Policy sets out the full detail.
No. There's no sign-up or login required, and we won't ask for your name, location or email address. Open the site, drop in your audio, and get back to the creative stuff.
Nothing. Narrly is free during early access, with nothing to pay and no card details to enter.
If anything here needs clarifying, please get in touch at hello@jamesmillner.dev.