Transcribe Files
Queue up any number of audio or video files, pick a mode once, and transcribe them in a batch.
The Transcribe Audio screen holds a queue. Drop in any number of audio or video files, pick the mode once, and start — each file is transcribed in turn while you keep working.
How to Use
- Drag files onto the screen, or use Add Files.
- Pick a Viva Mode for the run.
- Press Start.
Each row carries its own state — waiting, the current phase, done, or failed with a retry. A finished row expands in place to show its transcript, so you can read one result while the rest are still running.
The Mode Picker
A Viva Mode already carries the transcription model, AI provider and prompt, so the mode picker is the only control this screen needs — there is no separate AI toggle or preset picker here.
The choice is scoped to this screen. Your current setup is snapshotted, the mode is applied for the run, and the snapshot is restored afterwards, so transcribing files never rewrites your dictation setup. The picker returns to the Default mode each time you open the screen.
Failures
A file that produces an empty transcript is marked failed rather than silently discarded, so a bad input is visible instead of looking like it worked. Any failed row can be retried on its own without re-running the queue.
Length Limits
Some cloud providers cap upload size. VivaDicta checks before uploading and tells you the limit in minutes, along with providers that have no cap — rather than surfacing the vendor's raw error. Local models have no limit.
Related
- YouTube — transcribing from a URL
- Viva Modes — what a mode carries
- Transcription Models — choosing local or cloud