Bring an existing recording into Studio.
Import from audio file processes a finished recording into a new creative. Choose whether to retain the spoken performance, build editable speech with AI voices, or focus on its background audio.
Before you import
- Prepare an .mp3 or .wav file up to 20 MB and check that it plays correctly.
- Confirm permission to process, edit, and redistribute the voices and assets in the recording.
- Decide whether the spoken content should remain a recording or become editable speech with a similar AI voice.
- Keep the project open until the job reports a result.
Choose the import strategy
| Section | Option in the dialog | Result to expect |
|---|---|---|
| Voice & script | Keep original voices | Transcribed text linked to recorded vocal clips. Editing the transcript does not change the recorded words. |
| Voice & script | Make the script editable with similar AI voices (recommended) | Speech reconstructed with suggested synthetic voices so you can revise the script and generate new audio. |
| Voice & script | Skip spoken audio | No vocal clips are created from the file. |
| Music & sound effects | Keep existing music and sound effects | Preserve the extracted background where processing can recover it. |
| Music & sound effects | Replace with suggested music and sound effects | Use suggested alternatives where available; review the result before relying on it. |
| Music & sound effects | Skip background audio | No background audio is added from the file. |
The default selection is Make the script editable with similar AI voices plus Keep existing music and sound effects. You cannot select both Skip spoken audio and Skip background audio, because that would leave nothing to import.
Run the import
- Choose Import → Audio File – Ad creative audio file in the Studio toolbar.
- Drag a file into the upload area, or click to browse. Once attached, play its waveform preview to verify the source.
- Choose one Voice & script option and one Music & sound effects option.
- Click Confirm. Studio uploads the file and opens a processing dialog.
- Wait for the new creative to be created and selected, then play it from start to finish.
Understand the result
The new creative can contain a transcript, generated or PreRecorded vocal clips, and processed background material, depending on the options and source file. With similar AI voices, Studio may run Enhance after import to add expressive cues to generated speech. A speaker with too little usable speech may remain recorded. Transcript text describes a recorded performance; changing it does not automatically change words already spoken in the file.
- Check transcript boundaries and speaker changes.
- Listen for separation artifacts and verify any suggested background assets.
- Confirm the creative duration and clip positions.
- For editable vocal clips, check the selected voices and regenerate speech after copy changes.
- Keep the imported creative as a reference before making extensive edits.
If processing takes longer
Import runs asynchronously. Do not submit the same file repeatedly while a job is pending. If status does not advance, record the project, file name, approximate start time, and visible job state before contacting support.