How to Transcribe Audio to Text with AI (Free and Built-In Tools)

Before you start

• A clear audio or video recording (or a live speaker)
• One of: Microsoft 365 subscription, an iPhone on iOS 18 or later, a Google Pixel phone, or a Google account with Chrome
• Permission from the people being recorded

Method 1 Transcribe a recording in Microsoft Word for the web

1. 1

   Step 1: Open Word for the web

   Go to office.com or word.cloud.microsoft, sign in with the Microsoft account that has your Microsoft 365 subscription, and open a new blank document. Transcribe works in Edge or Chrome.

2. 2

   Step 2: Open Transcribe

   On the Home tab, click the arrow next to Dictate and choose Transcribe. The Transcribe pane opens on the right.

3. 3

   Step 3: Upload your file

   Click Upload audio and choose your recording. Supported formats are .mp3, .wav, .m4a and .mp4, up to 200 MB per file. Transcription takes roughly as long as the recording, or less.

4. 4

   Step 4: Review speakers and text

   The transcript appears in sections with timestamps and speaker labels (Speaker 1, Speaker 2). Click the pencil icon to fix words or rename speakers, and click a timestamp to replay that moment.

5. 5

   Step 5: Add it to your document

   Click Add to document and choose whether to include speakers and timestamps. Your audio file is saved in OneDrive.

Method 2 Transcribe on iPhone with Voice Memos

1. 1

   Step 1: Update your iPhone

   Voice Memos transcripts need iOS 18 or later and currently work for English recordings. Check in Settings > General > Software Update.

2. 2

   Step 2: Record or open a memo

   Open Voice Memos, tap the red button to record, and swipe up on the recording panel to see a live transcript (tap the speech-bubble icon).

3. 3

   Step 3: View the transcript

   For a saved recording, tap the three dots and choose View Transcript. Tap any word to jump to that point in the audio.

4. 4

   Step 4: Copy the text

   Tap Edit, then the speech-bubble icon, select the text and tap Copy to paste it into Notes, an email or a chatbot.

Method 3 Transcribe on a Google Pixel with Recorder

1. 1

   Step 1: Open Recorder

   Open the Recorder app (pre-installed on Pixel phones) and tap the red record button.

2. 2

   Step 2: Watch the transcript

   Tap Transcript to see words appear as people speak. It works offline, and Recorder labels different speakers.

3. 3

   Step 3: Search, summarise and export

   After recording, search the transcript for any word. On supported Pixel models, tap Summarise for key points. Use Share > Transcript to copy it or save it to Google Docs.

Method 4 Type live speech in Google Docs (free)

1. 1

   Step 1: Open Google Docs in Chrome

   Create a new document at docs.google.com using the Chrome browser.

2. 2

   Step 2: Start voice typing

   Click Tools > Voice typing, choose your language, and click the microphone. Speak clearly; words appear as you talk.

3. 3

   Step 3: Add punctuation

   Say “comma”, “full stop” or “period”, “question mark” and “new paragraph” (supported in English and some other languages).

Method 5 Clean up and summarise the transcript with AI

1. 1

   Step 1: Paste the transcript into a chatbot

   Open ChatGPT, Gemini or Copilot and paste the text (or upload the document).

2. 2

   Step 2: Ask for what you need

   Try: “Summarise this meeting in 5 bullet points”, “List every action point with the person responsible”, “Correct punctuation and remove filler words like um and uh, but don’t change the meaning”, or “Turn this interview into a Q&A article”.

3. 3

   Step 3: Check against the audio

   Spot-check quotes, names and numbers against the recording before you publish or send anything.

Troubleshooting

Frequently asked questions

Summary