Can I upload audio files to Wispr Flow for transcription?
Last updated: November 14, 2026
Available on: Mac, Windows, iOS, Android
No. Wispr Flow transcribes live speech from your microphone. You can't upload, attach, or import an existing audio or video file for transcription on any platform or plan — including Voice Memos, meeting recordings, .mp3 and .m4a files, or recordings stored in cloud services.
What Flow captures instead
Dictation: Requires microphone access; Mac also requires Accessibility permission, and the iOS keyboard requires Full Access — see Set up the Flow keyboard on iPhone. On desktop, you must be signed in and Flow must be ready; sign in again if prompted.
Live meetings: On Mac and Windows, Notetaker captures microphone and system audio during the meeting and turns its own recording into a refined transcript and summary — not a recording you supply. Mac also requires system-audio permission: grant it in the system prompt or System Settings when directed. If system-audio capture fails at the start, microphone recording continues.
Note: Notetaker uploads its recording for cloud processing and keeps the local audio after upload.
Retry a past dictation
Past dictations stay retrievable from History on Mac, Windows, iOS and Android. A retry re-transcribes Flow's own saved dictation audio, not a file you choose.
Mac and Windows: Retry failed or partial dictations from History, but not while dictation is active.
Android (History): Retries apply only to failed dictations. Open the ⋮ menu at the end of a card's timestamp row, where Copy, Retry, Report and Delete live. Retry appears only when the entry still has a saved recording, and while a retry runs Retry and Delete are grayed out, so the same recording can't be retried twice. If the recording is too long to retry, Flow tells you; if it contains no speech, the retry still succeeds and the entry is saved as an empty transcription. “Retry failed. Audio file not found.” means the linked recording is no longer on the device.
Android (error card): Retry on the dictation bubble's error card makes one fresh attempt. You can't start a new dictation while it runs — pressing to dictate shows “Still retrying your last dictation. Try again in a moment.”
Warning: Your retention setting decides whether that audio still exists. Where the options are normal storage, deletion after 24 hours, or Never store, the stricter of your choice and your organization's policy applies — except on desktop when your organization requires normal storage. Never store deletes existing local history and stops new transcripts and audio from being saved, so nothing is left to retry. On Android, reaching the device's history limit also removes the oldest entries and their recordings.
Import existing meeting transcripts
On Mac and Windows you can import meeting transcripts you already have — text only. Audio entries in an export are ignored, not transcribed. You must be signed in with Notetaker and the matching import source enabled, or the import card is hidden. Granola files and Otter ZIPs must be 50 MB or smaller.
Warning: Only one Granola or Otter import runs at a time. Choosing Granola again cancels an active Granola import so you can restart; choosing Otter never cancels an active import. Finish or cancel a blocking import before switching sources.
Open Settings → Notetaker and find Import your meeting notes at the bottom. The same flow opens from the Notetaker banner and Home's empty past-meetings list.
Choose Import, or Import again after a completed, failed or cancelled import.
Select your source, then follow its export steps below.
Otter export
Use an Otter bulk export, not loose .txt or .srt files.
Export as TXT with speaker names and timestamps using Otter's bulk-export instructions.
Select Choose a zip file, or drag the export's .zip or extracted folder onto the drop zone. The file chooser accepts only .zip files; folders must be dragged.
Summaries must end in _summary.txt and match a transcript's folder and base filename; _summary.srt and unpaired summaries aren't imported. Summaries over 256 KB are omitted while their transcripts still import. If a size limit is exceeded, export fewer conversations; if no transcripts can be read, re-export as TXT. Wait for one archive to finish reading before selecting another.
Granola export
Granola exports can include transcripts, notes and summaries.
Select a .csv or a .zip containing the export's CSV files.
Every CSV file must have an identical header line. Rows without a document ID or a readable date are skipped, and summary fields over 256 KB, notes over 2 MB and transcripts over 8 MB are omitted.
Finishing, cancelling and repeating an import
Keep Flow open during import. Meetings can appear before their transcripts, which keep arriving after the modal closes; the settings card shows remaining transcripts and the last completed import date. If authorization is required, choose Finish import to complete the remaining steps. An imported meeting without a summary still finishes right away — its Summary tab reads “This meeting came without a summary.”
Each export selection imports once and expires after 30 minutes. Reselect it to repeat an import, or after signing out, changing accounts or restarting Flow; a selection cannot be imported under a different account.
Cancelling keeps meetings already imported. To cancel and restart:
Select Cancel in the import modal or settings card.
Choose Yes, cancel to stop; Go back keeps the import running.
Reopen the import flow with Import again, then reselect your source and export.
FAQs
Will audio file upload be added?
We don't have a timeline to share. Submit a feature request at support.wisprflow.ai, which reaches our support team.