Import a recording
Outcome
Section titled “Outcome”A local audio or video file becomes a Synthezia session and begins the conversion and transcription workflow.
Prerequisites
Section titled “Prerequisites”- A local file you are authorized to process.
- A readable source file and an installed Whisper model, or a network connection to download the requested model.
- Select Upload Audio in the title bar, or press Cmd+O.
- Choose one file in the native picker.
- If Synthezia asks for a transcription model, select Download and continue and wait for the download to finish.
- Keep the session open while its status changes from Converting to Processing, then to Ready or Completed.
- Open the session information drawer to correct its title or add context while it processes.
Expected result
Section titled “Expected result”Synthezia converts the selected file to local 16 kHz mono WAV audio before running its local Whisper transcription path. A completed transcript is stored with the session.
Data boundary
Section titled “Data boundary”The selected source file is local. Downloading a missing model is a separate network action; an optional External API can change the boundary for later summaries or chat.
Common problems
Section titled “Common problems”- The picker does not accept the file: the current picker lists a bounded set of common extensions; it is not an unbounded FFmpeg compatibility promise.
- Conversion fails: retain the source file and error message. Codec support depends on the actual source and bundled FFmpeg build.
- A model dialog appears: download the required local transcription model, then the import resumes.