Three steps
- Add the audio. Drag in an MP3, M4A, WAV or video file, or record a meeting directly in Scrieb. It captures your microphone and system audio, so no bot joins the call.
- Transcribe locally. Whisper AI runs on your Mac or PC. Speaker identification runs alongside it, also on the device.
- Export. Click Export, pick DOCX, PDF or TXT. Open the DOCX in Word and edit it like any other document.
What the Word document looks like
Speaker names
Each turn starts with the speaker. Rename "Speaker 1" to a real name once and every occurrence updates.
Paragraphs, not a wall
Turns are split into readable paragraphs with punctuation, ready for quoting or editing.
Optional summary
Generate a summary or ask questions about the transcript before exporting. The AI chat runs locally as well.
Editable
Plain Word formatting, no locked fields. Works in Word, Google Docs, Pages and LibreOffice.
Which format for which job
| Format | Use it when | Notes |
|---|---|---|
| DOCX | You will edit, quote or send the transcript | Opens in Word and everything else. The default for interviews and client notes. |
| You need a fixed copy for a file or a client | Nothing can be changed by accident. Good for records. | |
| TXT | You will paste the text somewhere else | Analysis software, a CRM, a prompt. Smallest and most portable. |
All three come out of the same local pass, so the choice is only about what you do next — see audio to text for the formats Scrieb reads on the way in.
Formats, languages and recordings from your phone
| Input | Supported |
|---|---|
| Audio files | MP3, WAV, M4A, AAC, FLAC, OGG, Opus, WMA, AIFF |
| Phone recordings | iPhone Voice Memos (M4A), Android recorders (M4A, 3GP), WhatsApp voice notes (Opus) |
| Video files | MP4, MOV, M4V, MKV, WebM: the audio track is converted to text |
| Languages | About 100 spoken languages, including English, German, French and Spanish, detected from the speech |
| Output | DOCX for Word, PDF, TXT, Markdown |
You get editable text, not a locked transcript: correct names in the Document tab, then export. Recordings from mobile devices are moved to the computer first (AirDrop, USB or a cloud folder); see voice memo to text. Converting audio to text this way replaces a transcription service for most audio recordings: no upload, no per-minute price.
Compared with Word's own Transcribe
| Word Transcribe (Microsoft 365) | Scrieb | |
|---|---|---|
| Where the audio is processed | Microsoft's servers | Your computer |
| Monthly limit | Yes, upload minutes per month | None |
| Speaker labels | Yes | Yes, computed on the device |
| Record system audio | No | Yes, calls and meetings |
| Summary | With Copilot, in the cloud | Yes, local |
| Export | Into the open document | DOCX, PDF, TXT |
| Works offline | No | Yes |
Longer comparison: Microsoft Word Transcribe alternative.
Why offline matters for documents
A Word transcript usually ends up in a client file, a case folder or a research archive. If the audio went through a cloud service first, that service holds a copy and sits in your data-processing chain. With local transcription the only copies are the ones you make. For regulated work that removes the vendor question entirely; see local transcription and security.
Pricing
Free plan: 1 hour of recording per day with speaker labels and Word export. Pro: no minute limits, one licence for Mac and Windows.
Try it on your own recordings
Frequently asked questions
- How do I transcribe audio to text in Word?
- Microsoft Word for the web has a Transcribe button for Microsoft 365 subscribers: it uploads the recording to Microsoft and counts against a monthly allowance. To keep the recording local, transcribe it in Scrieb and export a DOCX, which opens in Word with speaker names in bold.
- How do I convert an audio file to a Word document?
- Drop the file into Scrieb, wait for the local transcription, then click Export and choose DOCX. Word, Google Docs, Pages and LibreOffice all open the result. The audio never leaves your computer.
- Can Scrieb transcribe directly into a Word document?
- Not live into an open document: Scrieb transcribes the recording first, then one click on Export → DOCX creates the Word file, with speaker names in bold and the summary if you want it.
- Does the Word document show who said what?
- Yes. Scrieb identifies speakers on your device and starts each turn with the speaker name, so a two-person interview reads as a dialogue rather than a wall of text.
- Can I export to PDF and plain text too?
- Yes. The same transcript exports as DOCX, PDF or TXT. PDF for sharing a fixed copy, TXT for pasting into analysis tools or an LLM.
- Which audio formats can I convert?
- MP3, M4A, WAV, FLAC, OGG and video files such as MP4 and MOV. No conversion step first.
- Is this different from Transcribe in Microsoft Word?
- Word's built-in Transcribe uploads your audio to Microsoft and counts against a monthly allowance in Microsoft 365. Scrieb does the same job on your own computer with no upload and no cap. Details on the Word Transcribe alternative page.
- How long does a one-hour recording take?
- A few minutes on a Mac with Apple Silicon or a PC with a modern GPU; about an hour on a laptop without a dedicated graphics card. It runs in the background.
- Is it free?
- Scrieb is free to try: 1 hour of recording per day, no card. Pro removes the limit and covers Mac and Windows with one licence.