Help Center / Start here

What is HoldSay, and what can you do with it?

Local voice tools for Mac and Windows.

HoldSay is a floating voice app for Mac and Windows. It turns speech into text wherever your active cursor is, so you can speak directly into the app or field you are using. It can also record your voice and the voices of other people in a computer call without adding a meeting bot.

HoldSay treats dictation, call recording, and file transcription as one voice workflow rather than separate tools. Use the floating voice card for live dictation and recording, then find completed work in the HoldSay library. This means less switching between services and a more direct path from spoken information to text and useful notes.

The HoldSay app icon above the floating voice card in its light and dark dress

One floating voice card for dictation, recording and transcripts — in light and in dark.

Use one voice card for writing, calls, and files

You might be replying to an email, joining a video call, or working through a folder of interviews. HoldSay is designed for those situations without making you move between separate dictation, recording, and transcription services. Its floating card is the place for held key dictation and REC recording, while completed work is kept in the app library.

HoldSay on Mac showing the recordings library, a speaker transcript, and the floating voice card

The floating voice card stays beside the HoldSay library, where recordings and transcripts remain available for review.

For writing, place your cursor in the field where you want words to appear. This can be an email, document, message, or another text field. Hold the dictation shortcut, speak, then release it. HoldSay inserts the finished text at the cursor you chose, so you can continue editing in the app where you were already working.

HoldSay showing the Dictating status while a text field is active

The Dictating status appears while HoldSay listens for text to place at the active cursor.

For a computer call, HoldSay can capture both your microphone and the sound playing through your computer. In practical terms, it can record your voice and the voices of people in a video meeting without adding a bot as a participant. Use the REC control on the floating card when you need a recording rather than text inserted into another app.

HoldSay recording microphone and computer audio on the floating voice card

The source label confirms that HoldSay is recording your microphone and the sound from the computer.

For material you already have, drop an audio file, video file, or folder into HoldSay. The app creates a transcript from the source, lets you replay the source, and can create a structured summary. This is useful when the conversation happened earlier and you need searchable notes instead of another live recording.

The Finder context menu on an audio file with Transcribe with HoldSay selected

HoldSay also lives in Finder and File Explorer: right-click an audio or video file and choose Transcribe with HoldSay.

This works without opening HoldSay first: right-click an audio or video file in Finder on Mac or File Explorer on Windows and choose Transcribe with HoldSay.

Start with a short dictation you can verify

The clearest first test is a short sentence in a text field you can see. It confirms that the shortcut reaches HoldSay, that your microphone is working, and that text is going to the intended location. The default shortcut is Right Option on Mac and Right Ctrl on Windows.

  1. Open an app with a text field, such as a draft email or document, and click where the text should be inserted. Leave the cursor in that field.
  2. Hold the default dictation shortcut. The floating card handles this held key dictation, so speak while you are holding the key.
  3. Release the shortcut when you finish the sentence. HoldSay places the finished text at the cursor in the text field.
HoldSay placing a spoken email into Gmail at the active cursor

A finished dictation is placed where the cursor is already active.

Read the inserted sentence before moving on. If it appears in the field you selected, the basic dictation workflow is working. You can correct the text in that app as you normally would.

If your first task is a meeting, make a short test call and use REC on the floating card. The useful result to check is that both sides of the call can be replayed: your microphone input and the computer audio from the other participant. If your first task is an archive, add one familiar audio or video file and compare the transcript with the source while replaying it. Familiar material makes it easier to recognize names, terms, and places that may need attention.

Understand what stays on your computer

HoldSay is a local desktop app. Once the required local models are ready, speech recognition and summaries run on your Mac or Windows PC. Your working audio and notes are kept in a library on that computer rather than being sent to a cloud transcription service for processing.

This arrangement matters when you want direct control over the files that support your work. You can open, back up, and move the ordinary files yourself. Local file library export is optional, so you do not need to export a library simply to use HoldSay.

Understand what stays on your computer in the HoldSay app, with Local file library highlighted

Open the current local library directly from Settings.

Every recording in the library also has its own folder of ordinary files. To see it, right-click a recording and choose Show in Finder.

The context menu on a HoldSay recording with Show in Finder highlighted

Show in Finder opens the folder behind any recording in the library.

A recording folder in Finder containing the audio, transcript, speakers, outline and metadata files

The recording's audio, transcript, speakers and outline are kept as plain files you can open, copy or back up.

On Mac and Windows, HoldSay can separate a recording into speakers and remember recurring voices. This makes recurring interviews and meetings easier to review because familiar names can carry into later notes.

HoldSay uses a device license rather than a cloud workspace account. This fits the local library approach while still allowing the app to perform the required license checks.

Find completed notes and choose the useful view

Use History to return to completed notes after a dictation, recording, or file transcription. It is one of HoldSay's main areas alongside Dictation, Files, and Settings. Keeping completed work in History gives you one place to review results instead of searching for each task where it began.

Find completed notes and choose the useful view in the HoldSay app, with Summary highlighted

Summary turns a long conversation into a shorter structure for review.

A completed note can show different views when they apply to its source. Raw is useful when you need the original recognized text. Speakers is useful for recordings where speaker separation applies. Polish gives you a refined view when that view is available, and Summary helps you review the structured result without reading the full transcript first.

Common questions

Is HoldSay a dictation app or a meeting recorder?

It is both. The same floating voice card handles cursor based dictation, meeting recording, and file transcription, and completed results are available in the local library.

Does HoldSay require a meeting bot?

No. HoldSay records the microphone and computer audio on the machine you are using, so it does not add a participant bot to the call.

Does HoldSay need an account?

HoldSay uses a device license rather than a cloud workspace account. A network connection is still needed for initial model downloads, license checks, and updates.

If a label on this page does not match the app, email support@holdsay.com. Include Mac or Windows and the HoldSay version shown in Settings.