Can I transcribe an existing audio or video file on my Mac?

Quick answer — November 1, 2026: Yes. On Mac, Dictop can transcribe audio and video files using drag and drop or Finder's “Transcribe with dictop” Services action. Enable saved history to retain the results; file transcription does not insert them automatically. Select a compatible on-device or cloud model, then review the transcript. File compatibility and results depend on the recording and selected model.

You do not need to replay a recording into the microphone to turn it into text. Import the file into Dictop to transcribe its audio directly.

Use it for recorded thoughts, an interview you are permitted to process, or narration from a video.

Choose the processing setup first

Before importing, select the speech model and language in Dictop. Enable Save transcriptions in Insertion settings so you can retrieve the result. If you are searching or editing history, also enable Save while searching or editing, or finish that activity before importing. For local audio processing, use an available on-device model and complete any required download. For cloud transcription, check your connection and credit balance.

If you enable AI refinement, choose a compatible option and decide what the edit should do. Start without summarization if you need a close transcript. Summarizing immediately can remove details you later want to inspect.

Choose a model suited to your recording. A model that works for short voice notes may be less suitable for noisy or specialized material.

Import and find the result

Drop the audio or video file into Dictop, or select it in Finder and use the Services transcription action. When processing finishes, look for the result in history. File import follows the history-saving controls; it does not force a saved entry when saving is disabled.

File transcription does not automatically paste the result into whichever app was focused. That difference is useful for longer material: you can inspect it before deciding what to copy into a document or share elsewhere.

Before processing a large collection, test a representative file in Dictop. File format and size support can affect whether a recording imports successfully.

Review a transcript differently from a summary

A transcript is recognized speech; an AI summary is an edited interpretation of that text. Check the transcript against the recording when accuracy matters, especially around names, numbers and unclear audio.

For recordings with several speakers, check whether you need speaker labels, timestamps or a specialized meeting transcript. File import does not necessarily provide those features.

Once the text is usable, Dictop's history tools let you search, edit and copy or share the version you need.

Choose where your recording is processed

An on-device speech model processes the audio locally. Cloud transcription sends it for the selected request; cloud refinement can also send text and prompts. Choose settings appropriate for the recording before importing it.

See supported media formats and transcript exports before choosing a workflow for your recordings.

Explore Dictop for Mac, read the Mac guide, and consult the privacy policy for processing details.