File Transcription

Process files in the queue#

1. Choose engine and model

   Use the global default or override the engine and model for file transcription only.

2. Add files

   Drop one or more audio or video files onto the view, or open them with Select files.

3. Follow the queue

   Files are processed sequentially. Cancel an individual job or the full queue; completed jobs remain available for copying and export.

4. Export the result

   Copy the text or save it as TXT. SRT and WebVTT are available only when the selected engine returns timestamped segments.

Formats and export limits#

Import: WAV, MP3, M4A, AAC, OGG, FLAC, WMA, MP4, MKV, AVI, MOV, WebM

Manual export: TXT is available for every completed result. SRT and WebVTT require real timestamped segments; an engine that returns plain text only cannot generate subtitles.

Watch folder: Automated output supports Markdown, TXT, SRT, and VTT. Subtitle formats still require timestamped segments.

Language and post-processing#

The manual queue uses the task, ordered language hints, number formatting, dictionary, and globally configured post-processing from dictation settings. You can override the engine and model in this view. App and website workflows are not selected automatically for queued files.

Watch Folder#

Choose an input folder and optionally a separate output folder. New supported files are processed automatically. Set the language, engine, and model, start watching with the app, and optionally delete each source file after successful processing.